Technical Guide to Network Video Surveillance (PDF)

Technical guide to network video Technologies and factors to consider for successful deployment of network video surveillance. November 2024. 2 Introduction We created this Technical guide to network video to support you in your work with Axis network video surveillance systems and to help you keep up with the changing technology landscape. Our guide is intended as a comprehensive resource for anyone involved in developing, implementing, and maintaining Axis video surveillance systems. It provides a complete overview of network video surveillance, and covers recent developments in cloud and video management systems (VMS). We hope you find Technical guide to network video useful! 3 Table of Contents 1. Network video: overview, benefits, and applications 9 1.1 Overview of a network video system 9 1.2 Benefits 10 1.3 Applications 12 2. Network cameras 15 2.1 What is a network camera? 15 2.1.1 AXIS Camera Application Platform (ACAP) 18 2.1.2 Application programming interface 18 2.1.3 ONVIF 18 2.2 Camera features for handling difficult scenes 19 2.2.1 WDR 19 2.2.2 Lightfinder technology 19 2.2.3 Day/night functionality 20 2.2.4 Built-in IR illumination 20 2.2.5 OptimizedIR 21 2.2.6 Lenses with low f-number 21 2.2.7 Automatic iris control 21 2.2.8 The right resolution 22 2.2.9 Thermal imaging 22 2.2.10 Image stabilization 23 2.3 Camera features for ease of installation 24 2.3.1 Outdoor-ready 24 2.3.2 Focused at delivery 24 2.3.3 Remote focus and zoom 24 2.3.4 Remote back focus 24 2.3.5 3-axis camera angle adjustment 24 2.3.6 Corridor format 25 2.3.7 Straighten image 25 2.3.8 Pixel counter 25 2.4 Types of network cameras 26 2.4.1 Box cameras 26 2.4.2 Bullet cameras 27 2.4.3 Dome cameras 27 2.4.3.1 Onboard dome cameras 28 2.4.3.2 Panoramic cameras 29 2.4.4 Modular cameras 31 4 2.4.5 PTZ cameras 34 2.4.5.1 Positioning cameras 37 2.4.5.2 Sharpdome technology 37 2.4.6 Thermal cameras 39 2.4.7 Explosion-protected cameras 42 3. Camera elements 45 3.1 Light sensitivity 45 3.2 Lenses 46 3.2.1 Lens types 46 3.2.2 Field of view 47 3.2.3 F-number and exposure 49 3.2.4 Iris types 49 3.2.5 Depth of field 50 3.2.6 Matching lens and sensor 51 3.2.7 Lens types in surveillance 52 3.2.8 Focusing 52 3.2.9 PTRZ precision 54 3.3 Removable IR-cut filter (day/night functionality) 54 3.4 Image sensors 57 3.5 Exposure control 58 3.5.1 Exposure priority 58 3.5.2 Exposure zones 58 3.5.3 Backlight compensation 59 3.5.4 Dynamic range 59 3.5.5 WDR imaging 60 3.6 Active tampering alarm 61 3.7 Audio detection 62 4. Video encoders 63 4.1 What is a video encoder? 63 4.1.1 Video encoder components and considerations 64 4.1.2 Event management and video analytics 64 4.2 Video encoders with analog PTZ cameras 65 4.3 De-interlacing techniques 65 4.4 Video decoders 66 5. Video resolutions 69 5.1 Megapixel resolutions 69 5.2 High-definition television (HDTV) resolutions 69 5.3 Ultra-HD resolutions 70 5 5.4 Aspect ratios 71 6. Video compression 73 6.1 Compression basics 73 6.1.1 Video codec 73 6.1.2 Image compression vs. video compression 74 6.2 Compression formats 76 6.2.1 H.264 77 6.2.2 H.265 77 6.2.3 Zipstream for H.264 and H.265 78 6.2.4 Motion JPEG 79 6.2.5 JPEG 80 6.2.6 MPEG-4 80 6.3 Average, variable, and maximum bitrates 80 6.4 Comparing standards 81 7. Audio 83 7.1 Audio capture 83 7.1.1 Devices and technology for audio-in connectivity 83 7.1.2 Audio capturing enhancements 85 7.1.3 Audio analytics 85 7.2 Audio output 85 7.2.1 Audio output in video surveillance 85 7.2.2 Network audio systems for public address 86 7.3 Audio communication modes 88 7.4 Audio codecs 88 7.5 Synchronization of audio and video 89 8. Radar 91 8.1 A non-visual technology that complements video surveillance 91 8.2 Radar devices 91 8.3 Radars in a surveillance system 92 9. Access control 93 9.1 What is access control? 93 9.2 Why IP in access control? 94 9.3 Access control components 95 9.3.1 Network door controllers 95 9.3.2 Readers 96 9.3.3 Network I/O relay modules 96 9.3.4 Access management systems 96 6 10. Network intercoms 97 10.1 Multifunctional communication devices for increased security 97 11. Wearables 99 11.1 The purpose of body worn cameras 99 11.2 Body worn cameras 100 11.3 Axis body worn solution 101 11.4 In-vehicle solution 103 12. Video analytics 105 12.1 Analytics for a smarter, safer world 105 12.1.1 Efficient monitoring 106 12.1.2 Efficient search 106 12.1.3 Efficient operations 106 12.1.4 Improved insights 107 12.1.5 Privacy 107 12.2 What are the benefits of video analytics? 107 12.3 Image usability – the foundation for great analytics performance 108 12.4 System architecture – where is the video analyzed? 108 12.5 Artificial intelligence 109 12.6 AXIS Camera Application Platform 110 13. Network technologies 111 13.1 Local area networks and Ethernet 111 13.1.1 Types of Ethernet networks 112 13.1.2 Connecting network devices and network switch 112 13.1.3 Power over Ethernet 113 13.2 Sending data over the internet 116 13.2.1 IP addressing 117 13.2.1.1 IPv4 addresses 117 13.2.1.2 IPv6 addresses 119 13.2.2 Data transport protocols for network video 119 13.2.3 SIP 121 13.3 VLANs 122 13.4 Quality of Service 123 14. System protection 125 14.1 Network protection 125 14.1.1 IEEE 802.1X 125 14.1.2 HTTPS (HTTP over TLS) 126 14.1.3 Trusted public key infrastructure (PKI) 127 7 14.1.4 NTS Network Time Protocol 127 14.1.5 Network isolation 127 14.2 Cybersecurity standards 127 14.2.1 ETSI EN 303 645 127 14.2.2 FIPS (Federal Information Processing Standard) 140 128 14.2.3 Common Criteria (CC) 128 14.3 Device protection 128 14.3.1 Built-in cybersecurity platform 128 14.3.2 User account management 129 14.3.3 IP address filtering 130 14.3.4 Keeping device software up to date 130 14.4 VMS protection 130 14.5 Physical protection 131 15. Wireless technologies 133 15.1 802.11 WLAN standards 133 15.2 WLAN security 134 15.2.1 WPA2™1 134 15.3 Wireless bridges 135 15.4 Wireless mesh network 135 15.5 4G and 5G networks 135 15.6 Z-Wave® 135 15.7 Zigbee® 136 16. Video management systems 137 16.1 Types of video management solutions 137 16.1.1 Cloud-based solutions 137 16.1.1.1 AXIS Camera Station Edge 138 16.1.2 Private networks 139 16.1.2.1 AXIS Camera Station Pro 140 16.1.3 Customized solutions from Axis partners 142 16.2 System features 142 16.3 Integrated systems 143 16.3.1 Access control 143 16.3.2 Point of sale 143 16.3.3 Building management 144 16.3.4 Industrial control systems 144 16.3.5 RFID 145 17. Cloud technologies 147 17.1 Moving to the cloud 147 8 17.2 Cloud setups 147 17.3 Axis Cloud Connect 148 18. System design considerations 149 18.1 Selecting a camera 149 18.1.1 Types of cameras 149 18.1.2 Image quality 150 18.1.3 Resolution 151 18.1.4 Compression 153 18.1.5 Networking functionality 154 18.1.6 Other functionalities 155 18.2 Installing a network camera 156 18.2.1 Surveillance objective 156 18.2.2 Handling challenging light 156 18.2.3 Lens selection 156 18.3 Physical protection of the camera 157 18.3.1 Protection and ratings 157 18.3.2 External housings 158 18.3.2.1 Positioning a fixed camera in a housing 159 18.3.3 Transparent domes 159 18.3.4 Vandal and tampering protection 160 18.3.5 Types of mounts 161 18.4 Bandwidth and storage considerations 164 18.4.1 Bits or bytes? 164 18.4.2 Bandwidth requirements 167 18.4.3 Storage requirements 167 18.4.4 System configurations 168 18.4.5 Edge storage 170 18.4.6 Server-based storage 172 18.4.7 NAS and SAN 172 18.4.8 Redundancy storage 174 19. Tools and resources 177 19.1 Find and compare products 177 19.2 Plan and design sites 177 19.3 Install and manage systems 178 20. Axis Communications Academy 179 20.1 Building expertise for smarter business 179 Chapter 1 — Network video: overview, benefits, and applications 9 1. Network video: overview, benefits, and applications A network video system uses standard IP-based networks for transporting video and audio. Digitized video and audio streams are sent over wired or wireless IP networks, enabling video monitoring and recording from anywhere on the network. With a multitude of advanced functionalities, network video has a lot to offer in security surveillance. The high-quality video, scalability, and built-in intelligence of network video enhance security personnel’s ability to protect people, property, and assets. 1.1 Overview of a network video system Network video, often also called IP-based video surveillance or IP surveillance as applied in the security industry, uses a wired or wireless IP network as the backbone for transporting digital video, audio, and other data. The network also carries power to the network video devices through Power over Ethernet (PoE) technology. A network video system allows video to be monitored and recorded from anywhere on the network, whether a local area network (LAN) or a wide area network (WAN) such as the internet. 10 Chapter 1 — Network video: overview, benefits, and applications Figure 1.1a A network video system with network cameras (1), analog cameras (2) connected through video encoders (3), and video management software (4). Other components, including the network, storage, and servers, are all standard IT equipment. Remote access is possible from a computer or a mobile device (5). The core components of a network video system are the network camera, the video encoder (used to connect analog cameras to an IP network), the network, the server and storage, and video management software (VMS). As network cameras and video encoders are computer-based, they have capabilities that cannot be matched by an analog CCTV camera. The network, the server, and storage components are all common off-the-shelf equipment — one of the main benefits of network video. Other components include accessories, such as mounts, PoE midspans, and joysticks. Each network video component is covered in more detail in other chapters. 1.2 Benefits A fully digital, network video surveillance system provides a host of advanced functionalities. Network video comes with high image quality, remote accessibility, video analytics, easy integration possibilities, scalability, flexibility, cost-effectiveness, and secure communication. > High image quality. In video surveillance, high image quality is essential to clearly capture an incident and to identify the persons or objects involved. With progressive scan and HDTV/ megapixel technologies, a network camera can deliver high image quality and high resolution. For more on image quality, see chapters 2, 3, and 6. Chapter 1 — Network video: overview, benefits, and applications 11 > Remote accessibility. Network cameras and video encoders can be configured and accessed remotely, enabling multiple, authorized users to view live and recorded video at any time and from virtually any networked location. This is advantageous if you need a third party, such as an alarm monitoring center or law enforcement, to also have access to the video. > Video analytics. In network video, a camera can be much more than just a source of video. Analytics in the camera are used to extract the useful information from massive amounts of video and trigger instant, automatic actions when necessary. This may include notifying security staff or starting a video recording upon detection of specific events. Analytics that use deep learning algorithms for object detection and classification can extract very detailed information about people and vehicles in a scene, enabling extremely accurate and efficient searches. For more on analytics, see chapter 12. > Easy, future-proof integration. Network video products based on open standards can be easily integrated into a wide array of video management systems. Video from a network camera can also be integrated into other systems, such as point-of-sales, access control, or a building management system. For more on integrated systems, see chapter 16.3. > Scalability and flexibility. A network video system can grow with the user’s needs. Video products and other types of applications share the same wired or wireless IP network for communicating data. Video, audio, PTZ and I/O commands, other data, and power are carried over the same cable, and any number of devices can be added to the system with no significant or costly changes to the network infrastructure. Devices can be placed and networked in virtually any location, and the system can be as open or as closed as desired. Since a network video system is based on standard IT equipment and protocols it can benefit from those technologies as the system grows. For instance, video can be stored on redundant servers placed in separate locations for greater reliability and security, and tools can be used for automatic load sharing, network management, and system maintenance. > Cost-effectiveness. An IP surveillance system typically has a low total cost of ownership. The network infrastructure is often already in place and used for other applications within an organization, so a network video application can piggyback off the existing infrastructure. Management and equipment costs are also low, since back-end applications and storage run on industry-standard, open systems-based servers, and not on proprietary hardware such as DVRs in the case of an analog CCTV system. Many network video devices are powered by Power over Ethernet (PoE) technology, which provides power through the same Ethernet cable that transports the data (video). PoE keeps installation costs down and facilitates backup. For more on PoE, see chapter 13. > Secure communication. Network video devices, as well as their video streams, can be secured in many ways. These include user name and password authentication, IP address filtering, 12 Chapter 1 — Network video: overview, benefits, and applications authentication using IEEE 802.1X, data encryption using HTTPS (SSL/TLS), and by using multiple user access levels. For more on network security, see chapter 14. > Increased security. Axis devices are safeguarded by the hardware-based cybersecurity platform Axis Edge Vault, which minimizes a device’s exposure to cybersecurity risks and enables it to be a trusted and reliable unit within the network. Axis Edge Vault enables cybersecurity features such as signed OS (guaranteeing that AXIS OS has not been compromised), secure boot (ensuring that unauthenticated or altered code is rejected during the boot process), and signed video (verifying video authenticity through a signature in the video stream). Existing analog video installations can migrate to a network video system and take advantage of some of the digital benefits with the help of video encoders and devices such as Ethernet-over- coax adapters, which make use of legacy coax cables. For more on video encoders and decoders, see chapter 4. 1.3 Applications Network video can be used in an almost unlimited number of applications. Most uses fall under security surveillance or remote monitoring of people, places, property, and operations. Increasingly, network video is also being used to improve business efficiency, as the number of video analytics applications grows. The following are some typical application possibilities in key industry segments. > Retail. Network video systems in retail stores can significantly reduce theft, improve staff security, and optimize store management. > Transportation. Network video helps to protect passengers, staff, and assets in all types of transport. > Banking and finance. Network video systems enable a bank to efficiently monitor its headquarters, branch offices, and ATM machines from a central location. > City surveillance. Network video is one of the most useful tools for fighting crime and protecting citizens. It can be used to detect and deter. > Education. From day-care centers to universities, network video systems help to deter vandalism and increase the safety of staff and students. > Government. Network video can be used by law enforcement, the military, and border control. It is also an efficient means to secure all kinds of public buildings. > Healthcare. Network video enables hospitals and healthcare facilities to improve the overall safety and security of staff, patients, and visitors. Chapter 1 — Network video: overview, benefits, and applications 13 > Industrial. Network video is not only an efficient tool for securing perimeters and premises, it can also be used to monitor and increase efficiency in manufacturing lines, processes, and logistics systems. > Critical infrastructure. Whether a solar plant, an electrical substation, or a waste management facility, network video can help ensure safe, secure, and uninterrupted activity. Production data from remote sites can be enhanced with visual information. 14 Chapter 1 — Network video: overview, benefits, and applications Chapter 2 — Network cameras 15 2. Network cameras Network cameras, or IP cameras, offer a wide variety of features and capabilities to meet the requirements of almost any surveillance system. This chapter provides a description of what a network camera is, the options and features it may have, and the various types of cameras available: fixed cameras, PTZ (pan-tilt-zoom) cameras, modular cameras, thermal cameras, and explosion-protected cameras. 2.1 What is a network camera? A network camera, also known as an IP camera, is used primarily to send video/audio over an IP network such as a local area network (LAN) or the internet. A network camera enables live viewing and/or recording, either continuously, at scheduled times, on request, or when triggered by an event. Video can be saved locally and/or at a remote location, and authorized access to video can be made wherever there is access to an IP network. Figure 2.1a A network camera (1) connects to a network switch (2) and video can be accessed (3) over the network. A network camera can be described as a camera and computer combined in a single unit. The main components include a lens, an image sensor, one or more processors, and memory. The processors are used for image processing, compression, video analysis, and networking functionalities. The memory is used mainly for storing the device software, but also for storing video. 16 Chapter 2 — Network cameras Like a computer, the camera has its own IP address, is connected directly to a wired or wireless network, and can be placed wherever there is a network connection. This differs from a web camera, which can only operate when connected to a PC. In addition to capturing video, network cameras provide event management that enables you to program automatic responses to predefined events. Typical responses include sending live video and email alerts, activating devices such as doors and lights, and initiating video recordings. By recording only when something specific happens in the scene, you make more efficient use of network bandwidth and storage space. Network cameras support video analytics, which automatically extract what is important in the video stream and uses that to provide actionable insights. Analytics that are embedded in the camera (edge analytics) come with several advantages, such as reduced network transmission needs and low latency for scenarios that require a quick response. Axis Scene Intelligence technology combines advanced image processing technology with analytics on the edge and deep learning to create a superior foundation for consistent analytics performance. It also enables the camera to adapt automatically when circumstances change. Many network cameras have support for audio. They can have audio input/output ports, or the possibility to connect audio devices through portcast technology or edge-to-edge technology (both of which provide seemingly camera-integrated audio). Network cameras also often offer input/ output (I/O) ports that enable connections to other external devices such as motion sensors, door lock relays, and audio/visual alerters. Most network video cameras are powered by Power over Ethernet (PoE) technology, which means that power is supplied through the network cable. They also have a memory card slot for local storage of recordings. Chapter 2 — Network cameras 17 Figure 2.1b Front, underside, and back of a typical network camera. 1. Zoom puller 2. Internal microphone 3. P-Iris lens 4. Focus puller 5. Serial port 6. I/O terminal block 7. Power connector 8. Iris connector 9. Memory card slot 10. Audio in 11. Audio out 12. Network connector 18 Chapter 2 — Network cameras Axis cameras are installed on a network through AXIS Camera Station Edge or AXIS Camera Station Pro, video management systems (VMS) that automatically detect and configure the camera. Advanced customers with large installations can use AXIS Device Manager to configure the cameras. It is also possible to access a camera directly through its built-in web pages by entering its IP address in a browser. Configuration concerns, for example, user access, camera settings, resolution, frame rate, compression format (H.265/H.264/Motion JPEG), as well as rules for events. Axis cameras also support a host of accessories that extend their abilities. For example, cameras can be connected to a fiber optic network using a media converter switch, or to coax cables using an Ethernet-over-coax adapter with support for PoE. 2.1.1 AXIS Camera Application Platform (ACAP) Most Axis cameras (and speakers, intercoms, and radar products) are supported by AXIS Camera Application Platform (ACAP). ACAP is an open application platform for a broad range of industries and use cases and it enables analytics applications, accessible from Axis website or from thirdparty suppliers. These applications can be downloaded and installed on the devices. ACAP also makes it possible to develop customized applications, and run applications on the edge (completely or partially) by combining advanced edge analytics and cloud or server-based technologies. For more on ACAP, see www.axis.com/developer-community/acap 2.1.2 Application programming interface All Axis network video products have an application programming interface (API) called VAPIX®. VAPIX enables developers to easily integrate Axis video products and their built-in functionalities into other software solutions. 2.1.3 ONVIF Most Axis network video products are ONVIF conformant. ONVIF®1 is a global, open industry forum founded by Axis, Bosch, and Sony in 2008, and its aim is to standardize the network interface of network video and access control products of different manufacturers to ensure greater interoperability. It gives users the flexibility to use ONVIF conformant products from different manufacturers in a multi-vendor, IP-based physical security system. ONVIF is today endorsed by the majority of the world’s largest manufacturers of IP-based physical security products, and has more than 500 member companies. For more information, visit www.onvif.org 1ONVIF is a trademark of ONVIF, Inc. Chapter 2 — Network cameras 19 2.2 Camera features for handling difficult scenes The video quality of security cameras may be negatively affected by challenging weather conditions or low or wide-ranging light levels. This chapter lists camera factors and features that impact the camera's ability to handle difficult scenes. 2.2.1 WDR When the light level changes in the scene, an Axis camera automatically adjusts to ensure optimal exposure. For challenging situations with scenes that contain both bright and darker parts, the WDR (wide dynamic range) option is recommended. This is enabled by default and is the best way to use Axis cameras because it does not need adjustments over time. WDR often enables the camera to combine short exposure times in the bright parts of the scene with long exposure times in the dark areas. The resulting image is correctly exposed in all areas. For more details on WDR imaging, see section 3.5.5. 2.2.2 Lightfinder technology Surveillance video with color greatly enhances the possibility to effectively identify people, vehicles, and incidents. Cameras with Axis Lightfinder technology have extreme light sensitivity, and can deliver day-mode color images in as little light as 0.08 lux, or lower. This is achieved through the optimal selection of image sensor and lens, Axis image-processing know-how, and in- house ASIC chip development. As these building blocks of Lightfinder regularly improve, Lightfinder, too, is constantly evolving. Lightfinder 2.0, which is suitable even for cameras with resolutions up to 4K, represents a step change in this evolution, with increased light sensitivity, a more life-like color reproduction, and customized tuning for advanced users. 20 Chapter 2 — Network cameras Figure 2.2a Left: a sharp, bright color image delivered by a Lightfinder 2.0 camera, even though the light intensity was only 0.05 lux under the bridge. Right: a snapshot of the same scene manipulated to visualize how the scene appeared to the human eye. 2.2.3 Day/night functionality A network camera with day/night functionality has an automatically removable infrared-cut filter. The filter is on during daytime, enabling the camera to produce colors as the human eye sees them. At night, the filter is removed to enable the camera to take advantage of near-infrared light and produce good quality black and white images. This is one way of extending a network camera’s usefulness in low-light conditions. Figure 2.2b Left: day mode. Right: night mode. 2.2.4 Built-in IR illumination In low light or complete darkness, built-in infrared (IR) LEDs in a camera (or a separately installed IR illuminator) will increase the camera’s ability to use near-infrared light to deliver quality black and white images. Near-infrared light from the moon, street lamps, or IR illuminators is not visible to the human eye, but a camera’s image sensor detects it. Chapter 2 — Network cameras 21 The built-in IR LEDs in Axis cameras can be adjusted to match the viewing angle and can be activated automatically in darkness, upon an event, or upon request by a user. Figure 2.2c Left: night mode snapshot captured without illuminators (a small amount of light was admitted under a door in the left-hand corner of the room). Right: night mode snapshot captured using IR illuminators. 2.2.5 OptimizedIR Axis cameras with OptimizedIR provide a unique and powerful combination of camera intelligence and sophisticated LED technology using Axis most advanced camera-integrated IR solutions. Examples include a patented technology for assuring an even illumination in the camera’s variable field of view, extremely efficient heat management, and the use of long-range, high-quality LEDs that are fine tuned to the camera. OptimizedIR is in constant development, with new advanced features being added. 2.2.6 Lenses with low f-number Camera lenses with a lower f-number have a better light gathering ability. In general, the lower the f-number, the better its performance in low-light settings. Sometimes a higher f-number is preferable for handling certain types of lighting. A camera’s light sensitivity depends not only on its lens, but also on the image sensor and image processing. More details on lenses and image sensors are provided in Chapter 3. 2.2.7 Automatic iris control For scenes with changing light levels, an automatically adjustable iris (DC-iris, P-Iris, or i-CS lens) is recommended to provide the right level of exposure. Cameras with a P-Iris lens or i-CS lens have better iris control for optimal image quality in all lighting conditions. More details are covered in Chapter 3. 22 Chapter 2 — Network cameras 2.2.8 The right resolution A camera’s resolution is defined by the number of pixels on the image sensor. A camera with a megapixel sensor delivers images with one million pixels or more. A high resolution is generally desired. When using a wide view angle, a camera with higher resolution provides a wider area of coverage. When using a narrow view angle, a camera with higher resolution provides greater detail, which is useful in identifying people and objects. But as the surveillance industry has continued to move to higher resolutions, manufacturers have usually tried to keep the same sensor size to avoid the higher cost of using a larger sensor. This means that each pixel must be smaller, and smaller pixels are able to capture less light. By simply increasing the number of pixels in a sensor of the same size you get better resolution, but your image may also have lower quality, especially in a low-light scene. A camera with a sensor of around 4 megapixels generally strikes a balance between resolution and light sensitivity because it provides a large enough pixel size without having to use a larger, more expensive sensor. Cameras supporting HDTV 720p (1280x720 pixels), HDTV 1080p (1920x1080 pixels), WQHD (2560x1440 pixels), and 4K Ultra HD (3840x2160 pixels), which are approximately 1, 2, 4, and 8.5 megapixels, respectively, follow standards that guarantee full frame rate, high color fidelity and a 16:9 aspect ratio. 2.2.9 Thermal imaging Besides sunlight, artificial light, and near-infrared light, thermal radiation can also be used to generate images. A thermal camera requires no light source, but instead detects the thermal radiation emitted by any object warmer than absolute zero (0 K). Thermal cameras can be used to detect subjects in complete darkness, in smoke or fog, or when subjects are obscured by shadows or a complex background. Nor are such cameras blinded by strong lights. Thermal cameras are ideal for detection purposes and can be used to complement conventional cameras in enhancing the effectiveness of a surveillance system. For more information about thermal cameras, see section 2.4.6. Chapter 2 — Network cameras 23 Figure 2.2d Left: image from a conventional camera on a foggy scene. Right: image from a thermal camera on the same foggy scene. 2.2.10 Image stabilization A surveillance camera mounted in an exposed location, such as on a high pole or a street sign near a busy road, can be shaken by winds or passing traffic. This could blur the video, especially when a powerful zoom lens is used. Having cameras that are less sensitive to vibrations makes installation more flexible and allows for multiple mounting options, even though vibration shall be avoided if possible. Because stabilized video will contain comparatively less movement, it requires less bandwidth and storage resources than shaky video. Real-time image stabilization techniques can make the video output less sensitive to vibration and maintain image quality. > Optical image stabilization (OIS) usually relies on gyroscopes or accelerometers to detect and measure camera vibrations. This method is particularly useful with long focal lengths and works well also in low light conditions. The main disadvantage of an optical solution is the price. > Electronic image stabilization (EIS) relies on algorithms for modeling camera motion, which then are used to correct the images. This method is cost-efficient, but sometimes fails to distinguish between physical motion induced by vibrations and perceived motion caused by fast-moving objects in front of the camera. Axis employs a stabilization method that is a hybrid of the two techniques. The Axis feature is called electronic image stabilization (EIS), but combines advanced gyroscopes together with optimized algorithms to make a robust and reliable system. It covers a wide band of vibration frequencies and copes with high and low amplitudes. EIS from Axis can always distinguish between physically induced vibrations and perceived motion. The system performs very well even in poor lighting since it relies on gyroscopic information, rather than video content, for its motion 24 Chapter 2 — Network cameras calculations. For the same reason, the system can always distinguish between perceived motion and physically induced vibrations. 2.3 Camera features for ease of installation Axis cameras incorporate features that make the products easy to install and use, as well as more reliable, by minimizing installation errors. 2.3.1 Outdoor-ready Axis outdoor-ready products are prepared for outdoor installation out of the box. No separate housing is required and the products come with captive screws that will not fall out of their screw holes. The products are designed to run in a range of operating temperatures and offer protection against dust, rain, and snow. Some even meet military standards for operation in harsh climates. 2.3.2 Focused at delivery To make installation faster and simpler, Axis cameras with a fixed focal lens are ready-focused at the factory, eliminating the need to focus them during installation. This is possible because fixed focal cameras with a wide or mid-range field of view usually have a wide depth of field (the range in which objects both near and far are in focus). For more on focal length, f-numbers and depth of field, see chapter 3. 2.3.3 Remote focus and zoom A varifocal camera with remote focus and zoom eliminates the need to manually, on site, adjust the focus and the field of view when the camera is installed. Thanks to the camera’s lens motor, this can be done remotely from a computer on the network. 2.3.4 Remote back focus A CS-mount varifocal camera with remote back focus allows the focus to be fine-tuned remotely from a computer, by making tiny adjustments to the position of the image sensor. This functionality also works with optional lenses. 2.3.5 3-axis camera angle adjustment An Axis dome camera is designed with a 3-axis angle adjustment that allows the lens holder (comprising the lens and image sensor) to pan, tilt, and rotate. This allows the camera to be mounted on a wall or ceiling. Users can then easily adjust the camera’s direction and level the image. The flexibility of the camera adjustment, together with the ability to rotate the image using Chapter 2 — Network cameras 25 the camera web interface, also means it is possible to get a vertically oriented video stream (corridor format). Figure 2.3a 3-axis camera angle adjustment 2.3.6 Corridor format Corridor format enables a fixed camera to provide vertically oriented video. The vertical format optimizes the coverage of areas such as corridors, hallways, and aisles, maximizing image quality while minimizing bandwidth and storage requirements. It enables, for example, HDTV network cameras to deliver video with a 9:16 aspect ratio. With a dome camera, this is achieved by first rotating the lens 90° (or with a bullet or box camera, by rotating the entire camera), and then rotating the video image back 90° in the camera’s web page. 2.3.7 Straighten image Straighten image is an image processing tool which makes it possible to digitally correct slight mechanical skews in image rotation (around the longitudinal axis). The goal is to level the image with the horizon. 2.3.8 Pixel counter Axis pixel counter is a visual aid shaped as a frame with a corresponding counter to show the frame’s width and height. The pixel counter helps ensure that the video resolution has sufficient quality to meet goals such as facial identification, and also to verify that the resolution of an object fulfills regulatory or customer requirements. For details about recommended pixel densities for different surveillance purposes, see chapter 18.1.3. 26 Chapter 2 — Network cameras Figure 2.3b Axis pixel counter showing the pixel resolution of a face. 2.4 Types of network cameras Network cameras may be designed for indoor use only, or for both indoor and outdoor use. An outdoor-ready camera is supplied with an external, protective housing. For more on protection, see section 18.3. 2.4.1 Box cameras Figure 2.4a Box cameras. A box camera is a traditional surveillance camera type. Both the camera and its viewing direction are readily apparent, making this camera type the best choice for deterrence purposes. Most box cameras come with an interchangeable lens, which may be fixed, varifocal, or with motorized zoom. This type of camera is available for both indoor and outdoor use. Axis box cameras for outdoor use come in protective housings. Chapter 2 — Network cameras 27 2.4.2 Bullet cameras Figure 2.4b Bullet cameras. Bullet cameras have a small and slim design, compared with box cameras. They are outdoor-ready, which means that they can be placed both indoors and outdoors, without any additional protective enclosure. All bullet cameras from Axis are equipped with built-in IR light which enables surveillance in low light or complete darkness. As with box cameras, the camera’s viewing direction is obvious. On bullet cameras, however, it is not possible to switch lenses. 2.4.3 Dome cameras Figure 2.4c Dome cameras. Dome cameras consist of a fixed camera preinstalled in a small dome-shaped housing. This camera’s main benefit lies in its discreet, unobtrusive design. In addition, people in the camera’s field of view find it difficult to see in which direction the camera is pointing. There is a wide selection of accessories that enable even more discreet installation, such as black casings, recessed mounts, and smoked domes. The dome camera is also more resistant to tampering than other fixed cameras. It may come with a fixed, varifocal, or motorized zoom lens. Dome cameras are designed with different types and levels of protection, such as vandal and dust resistance, and IP66, IP67, and NEMA 4X ratings for outdoor installations. No external housing is required. These cameras are usually mounted on a wall, ceiling, or pole. 28 Chapter 2 — Network cameras 2.4.3.1 Onboard dome cameras Figure 2.4d The onboard camera category consists of modular cameras, dome cameras, and panoramic cameras. Onboard cameras are specially designed for surveillance on rolling stock, busses, and other vehicles. The conditions in onboard surveillance vary depending on the type of vehicle the camera is used in. In some usage areas, such as on trains, there are specific protocols and regulations to follow, and only products tested specifically for these environments should be used. However, common for all onboard surveillance is the need for a particularly rugged design. This means cameras must withstand harsh conditions such as vibrations, shocks, and bumps, as well as dust, water, and temperature fluctuations. An onboard camera’s discreet appearance is often combined with an active tampering alarm that helps detect and prevent attempts to redirect and defocus the camera. Chapter 2 — Network cameras 29 Figure 2.4e An onboard dome camera mounted in the ceiling of a bus. 2.4.3.2 Panoramic cameras Figure 2.4f Panoramic cameras: single-sensor, multisensor, multidirectional, and a solution combining a multidirectional camera with a PTZ camera. A panoramic camera is a dome camera that provides wide area coverage and excellent image detail at the same time, in an efficient one-camera installation. The camera can be single-sensor, multisensor, or multidirectional. A single-sensor panoramic camera has one wide-angle lens that gives a 360° fisheye view. The view can be transformed, either live or on recorded material, into various rectangular views, including panoramic, double panoramic, or quad-view format (simulating four different cameras). 30 Chapter 2 — Network cameras Figure 2.4g A single-sensor panoramic camera offers multiple viewing modes such as 360° overview, quad view, and panorama. A multisensor camera provides seamless 180° coverage with great detail and minimal distortion. Because it has multiple sensors, it combines wide coverage with high image quality and high pixel density. It uses a universal white balance setting and synchronized exposure for all the sensors. With its seamlessly stitched image, a multisensor camera also eliminates blind spots. Figure 2.4h 180° view from a multisensor camera. A multidirectional camera with 360° coverage is a panoramic camera with individually adjustable camera heads. It is ideal in, for example, retail stores, hallways, and warehouses. A solution that combines a multidirectional panoramic camera with a PTZ camera is especially useful in, for example, intersections. Chapter 2 — Network cameras 31 2.4.4 Modular cameras Figure 2.4i Modular cameras including main units and sensor units. Axis modular cameras are designed to provide a flexible and discreet installation. Blending in with the environment, they can be used practically anywhere. The sensor units can be installed in tight spaces, used in buses and police cars, integrated with machines such as ATMs, and placed at eye level at exits for optimal face captures. They are ideal for highly discreet video and audio surveillance and analytics applications. A modular camera is a divided camera concept, where the main unit can be installed up to 30 m (98 ft) from the sensor unit. The sensor unit contains the lens and the image sensor, and the main unit processes the image. This divided concept provides flexibility in the choice of hardware, as well as in installation. The main and sensor units can easily be exchanged or relocated after the initial installation and the detachable cables ensure easy maintenance or upgrading. A discreet installation also reduces the risk of tampering. 32 Chapter 2 — Network cameras Figure 2.4j AXIS F Modular Cameras series offers a range of high-performance modular cameras designed for indoor, outdoor, and onboard surveillance. The series includes two rugged main units, two UL-recognized barebone main units, multiple cable options and sensor units for a wide range of applications. Modular cameras are also cost-effective when you need multiple cameras installed in one area. There are one- and four-channel main units, and a four-channel main unit connects to up to four sensor units. Thanks to maximum cable lengths up to 30 m (98 ft), one main unit with four sensor units can cover both larger and smaller areas. Four connected sensor units also enable simultaneous, multiple video streaming and quad view streaming using a single IP address. One main unit, whether it supports one or four video channels, requires a single VMS license and one switch port for cost-effective surveillance. Chapter 2 — Network cameras 33 Figure 2.4k Quad-view streaming from a four-channel modular camera system in a bus. Figure 2.4l At left, a pinhole modular camera in an ATM application. At right, a sensor unit mounted on a wall, and on a glass panel. A wide range of sensor units with different form factors and different lens types, such as standard, varifocal, mini-dome, and pinhole are available. 34 Chapter 2 — Network cameras Figure 2.4m Designed to be placed near building exits, a height strip housing enables extremely discreet installation of pinhole sensors. The camera is positioned to look straight at a person’s face for more reliable identification. 2.4.5 PTZ cameras Figure 2.4n PTZ camera models with features such as: (from left) HDTV 1080p resolution with Sharpdome technology, active cooling, a solution combining a multi-sensor camera and a PTZ camera, and (far right) a combination of a visual and a thermal camera in one unit. A PTZ camera provides pan, tilt, and zoom functions (using manual or automatic control), enabling wide area coverage and great detail when zooming in. Axis PTZ cameras usually have the ability to pan 360°, to tilt 180° or 220°, and are often equipped with a zoom lens. A true zoom lens provides Chapter 2 — Network cameras 35 optical zoom that maintains image resolution, as opposed to digital zoom, which enlarges the image at the expense of image quality. In operations with live monitoring, PTZ cameras can be used to follow an object, and to zoom in for closer inspection. In unmanned operations, automatic guard tours on PTZ cameras can be used to monitor different areas of a scene. In guard tour mode, one PTZ camera can cover an area that would require many fixed cameras to do the same job. Figure 2.4o Wide view (left) and zoomed-in view (right) in an HDTV 1080p PTZ camera, making the text on the cargo ship readable at 1.6 km (1 mile). Figure 2.4p Wide view (left) and zoomed view (right) in an HDTV 1080p PTZ camera, allowing the license plate to be read at 275 m (900 ft.). Axis PTZ cameras can be equipped with a variety of intelligent functionalities, for example: > Guard tour (preset positions). PTZ cameras usually allow multiple (up to 100) preset positions to be programmed. Once these positions have been set in the camera, the operator can quickly move from one position to the next, simply by selecting a position. In guard tour mode, the camera can be programmed to automatically move from one preset position to the next in a pre-determined order or at random. Normally up to 20 guard tours can be configured and activated at different times of the day. > Guard tour (recorded). The tour recording functionality in PTZ cameras enables easy setup of an automatic tour using a device such as a joystick to record an operator’s pan-tilt-zoom 36 Chapter 2 — Network cameras movements and the time spent at each position. The recorded tour can then be activated at the touch of a button or at a scheduled time. > Autotracking. This video analytics application will automatically detect a moving person or vehicle and follow it within the camera’s area of coverage. Autotracking is particularly beneficial in unmanned video surveillance situations where the occasional presence of people or vehicles requires special attention. The functionality reduces the cost of a surveillance system substantially, as fewer cameras are needed to cover a scene. It also increases the effectiveness of the solution, as it allows a PTZ camera to record the parts of a scene where there is activity. > AXIS Radar Autotracking for PTZ. This software application uses motion data from Axis radar devices to find objects of interest on a site, and automatically controls the direction and zoom level of one or more PTZ cameras. The radar device acts as a complement to the PTZ camera, and the application enables visual confirmation of detected objects even outside of the camera’s current field of view. AXIS Radar Autotracking for PTZ also minimizes the need for manual, joystick control of the camera. > AXIS Perimeter Defender PTZ Autotracking. When installed on a PTZ camera, this application allows a fixed thermal or fixed visual camera running AXIS Perimeter Defender analytics to automatically steer the PTZ camera for close-up views of alarm objects in the fixed camera’s detection zone. The PTZ camera automatically adjusts the zoom level to keep in view all alarm objects, including new ones that appear in the fixed camera’s detection zone. > Advanced/active gatekeeper. Advanced gatekeeper enables a PTZ camera to pan, tilt, and zoom in to a preset position when motion is detected in a predefined area, and to then return to the home position after a set time. When this is combined with the ability to continue to track the detected object, the function is called active gatekeeper. Chapter 2 — Network cameras 37 2.4.5.1 Positioning cameras Figure 2.4q A visual positioning camera. Positioning cameras deliver the most comprehensive field of view possible: mounted on wall, poles, or columns, they provide a panoramic view of 360° and a ground-to-sky view of 135° or more. They offer both high-speed and ultra-slow, jerk-free pan and tilt movements. Positioning cameras are specially designed to be reliable, robust, and weatherproof. Axis offers both visual and thermal PTZ positioning cameras, as well as bispectral PTZ positioning cameras that combine visual and thermal images. 2.4.5.2 Sharpdome technology With Axis Sharpdome technology, a PTZ camera can see above its horizon. Sharpdome offers innovative mechanics that makes the entire dome rotate, in contrast to a conventional dome where the camera rotates inside the dome. The mechanics and the placement of the camera module together with the unique design of the outdoor dome enable the same optimal image sharpness and full scene fidelity in all pan and tilt positions. This provides clear identification of objects as much as 20° above the camera horizon. 38 Chapter 2 — Network cameras Figure 2.4r A camera with Sharpdome (left) and a conventional dome camera (right) Sharpdome includes the speed dry function, which helps to provide sharp images in rainy weather. When visibility through the dome is impaired by water drops, speed dry rotates the dome in alternating directions at high speed, effectively shaking off the water. Figure 2.4s Two snapshots of the same rainy scene, before shaking (left) and after shaking off the water (right) using speed dry. Chapter 2 — Network cameras 39 2.4.6 Thermal cameras Figure 2.4t Thermal cameras in bullet (left) and fixed box mounted on positioning unit (center) form factors. The bispectral PTZ camera (right) combines a visual and a thermal camera in one unit for mission-critical surveillance. Thermal cameras detect the thermal radiation (heat) that all objects with a non-zero temperature emit. With the ability to pick up small temperature differences and convert them into a visual image, these cameras can distinguish persons and vehicles at great distances. They perform even in complete darkness and regardless of lighting conditions, camouflaging, vegetation, difficult weather, or other conditions where a visual camera might be found lacking. Thermal cameras are widely used in perimeter protection systems. Live video from a thermal camera can detect activity around critical locations long before a visual camera has seen anything unusual. The thermal images are analyzed automatically, directly in the camera, and the security system can be set up to respond in various ways. It can trigger automatic audio alerts in loudspeakers to actively deter intruders, it can email alerts to security personnel, and can and pan and zoom the system’s visual cameras to capture and record video footage in which the intruders can be identified. Thermal images alone are typically not enough to identify individuals. This makes thermal cameras a valuable option for surveillance in locations where privacy is especially important. Thermal cameras are also installed to monitor the temperature of industrial processes. They can be used, for example, to find heat leaks in buildings, or to determine whether a vehicle was recently used. A thermometric camera is a thermal camera designed to monitor objects or processes to detect if temperatures rise above or fall below a set of user-defined limits. This type of camera is used in order to prevent damage, failure, fire, or other hazardous situations. 40 Chapter 2 — Network cameras Figure 2.4u Image from a thermometric camera. Different temperatures are visualized using a color palette. A thermal camera requires special optics, since regular glass will block the thermal radiation. Most thermal camera lenses are made of germanium, which enables infrared light and thermal radiation to pass through. How much or how far a thermal camera can “see” or detect depends on the lens. A wide-angle lens gives a thermal camera a wider field of view, but a shorter detection range than a telephoto lens, which provides a longer detection range with a narrower field of view. A thermal camera also requires a special image sensor. Sensors for thermal imaging can be broadly divided in two types: Uncooled thermal image sensors operate at or close to the ambient temperature, at 8–14 μm in the long-wave infrared range. Uncooled sensors are often based on microbolometer technology, and are smaller and more affordable than cooled image sensors. Cooled thermal image sensors are usually contained in a vacuum-sealed case and cooled to temperatures as low as -210 °C (-346 °F) to reduce noise created by their own thermal radiation at higher temperatures. It allows the sensors to operate in the mid-wave infrared band, approximately 3–5 μm, which provides better spatial resolution and higher thermal contrast since such sensors can distinguish smaller temperature differences and produce crisp, high resolution images. The disadvantages of such detectors are that they are bulky, expensive, energy-consuming and the coolers must be rebuilt every 8,000 to 10,000 hours. Chapter 2 — Network cameras 41 Figure 2.4v The spectrum of electromagnetic radiation. Axis thermal cameras work in the long– wavelength IR region (7) at about 8–14 μm. 1. X-rays 2. Ultraviolet light 3. Visible light 4. Near-infrared (NIR) radiation (approximately 0.75–1.4 μm) 5. Short-wavelength infrared (SWIR) radiation (1.4–3 μm) 6. Mid-wavelength infrared (MWIR) radiation (3–5 μm) 7. Long-wavelength infrared (LWIR) radiation (8–14 μm) — used by Axis thermal cameras 8. Far-infrared (FIR) radiation at approximately 15–1,000 μm 9. Microwave radiation 10. Radio/TV wavelengths 11. IR illumination 12. Axis thermal cameras A thermal sensor’s ability to detect very small differences in thermal radiation can be characterized by its NETD (noise equivalent temperature difference) value. In general, the smaller the NETD, the better the sensor. However, cameras should not be rated by comparison of NETD specifications only, due to the lack of a standardized measurement protocol. Products and technologies that can be used both for military and commercial purposes are called dual-use goods. Exports of such items are regulated in the international Wassenaar Arrangement from 1996, which, among other things, aims to promote transparency and greater responsibility in transfers of conventional arms, as well as dual-use goods and technologies. For a thermal camera to be freely exported, its maximum frame rate cannot exceed 9 frames per second (fps). Thermal cameras with a frame rate up to 60 fps can be sold in most 42 Wassenaar Arrangement member countries on the condition that the buyer is registered and can be traced and that the seller fulfills all country-specific export requirements, including license or license exception. Thermal cameras may never be sold to any embargoed or sanctioned destination. 42 Chapter 2 — Network cameras 2.4.7 Explosion-protected cameras Figure 2.4w Explosion-protected cameras. Left and middle: Fixed and PTZ cameras designed and certified for Zone/Division 1 hazardous areas. Right: Fixed camera designed and certified for the less potentially combustible Zone/Division 2 areas in hazardous locations. Explosion-protected cameras are ideal for health and safety applications as well as operational efficiency. You can use them to monitor operations and processes with remote visual verification of readings from gauges and sensors across your sites. They can help increase efficiency, maintain peak operability, and maximize uptime. Cameras used in hazardous locations can be integrated with a system of sensors and safe-area cameras already in place on an existing network. Thermometric cameras can monitor potential overheating of equipment, which enables informed decisions about the ongoing process, helping to keep production up-and-running and as efficient as possible. With explosion-protected cameras, you can improve the safety of employees and minimize unnecessary human exposure to hazardous areas. You can use analytics that monitor the use of personal protection equipment, such as hard hats, or that detect signs of smoke or fire in potentially combustible environments. Chapter 2 — Network cameras 43 Figure 2.4x Typical industry segments for explosion-protected equipment are processing industries, oil and gas installations, and grain handling and storage. Hazardous areas can, however, also be found in a wide range of other industries, such as chemical plants, pharmaceuticals, mining, and water and waste treatment. Explosion-protected network cameras offer many advantages over analog versions, the major ones being built-in intelligence, video analytics, recording and storage possibilities, superior image quality, and a modern, future-proof camera technology. When an electrical device is installed in a potentially combustible environment, such as in chemical processing plants, or near gas valves or vents, it must meet very specific safety standards. It is the potentially combustible environment that must be protected from ignition. Explosion-protected cameras are certified for use in hazardous locations because the cameras are designed to comply with an explosion-protection method, which could be, for example, containment or prevention. The term explosion-protected does not imply that the camera itself will withstand an external explosion. > Containment: These cameras utilize a heavy-duty enclosure to confine the energy of a potential internal combustion. In case of explosion caused by sparks or high temperatures in these cameras, the explosion will be limited to within the enclosure and not spread to the possibly flammable atmosphere outside of it. This type of explosion-protected camera can be certified for use in Zone/Division 1 hazardous areas, where there is a relatively large likelihood that the atmosphere is flammable due to its concentration of explosive vapors, gases, dust, or fibers. > Prevention: These cameras are designed so that they cannot provide sufficient energy to ignite a flammable atmosphere. They can be certified for use in Zone/Division 2 hazardous areas, which are less potentially combustible areas in a hazardous location. In hazardous locations, Zone/Division 2 areas are typically significantly larger than Zone/Division 1 areas. Cameras certified for Zone/Division 1 areas can also be used in Zone/Division 2 areas, but cameras specifically designed and certified for Zone/Division 2 areas are a more cost-efficient alternative in these areas. 44 Chapter 2 — Network cameras Figure 2.4y The right camera for the right type of hazardous area. 1. In Zone/Division 1 areas you must use a camera specifically certified for Zone/Division 1 areas. 2. In the large, less potentially combustible (Zone/Division 2) areas of a hazardous area site, it is also possible to use a substantially lighter and more cost-efficient camera certified for Zone/ Division 2 areas. Explosion-protected cameras must be certified according to the industry standards applicable in the country where the camera will be used. Chapter 3 — Camera elements 45 3. Camera elements Image quality and field of view may be the most important aspects of any camera. They are affected by a number of camera elements, including the camera’s light sensitivity, the lens it uses, the image sensor, and image- processing functionalities. 3.1 Light sensitivity A camera’s light sensitivity, often specified in lux, is the lowest illuminance level at which the camera produces an acceptable image. Illuminance is a measure of how much the incident light illuminates a surface, per area unit. The light sensitivity depends mainly on the lens and the image sensor, and the lower the specified lux value, the better the light sensitivity of the camera. Illuminance (lux) Light condition 100,000 Strong sunlight 10,000 Full daylight 500 Office light 100 Family living room 10 Candle at a distance of 30 cm (1 ft.) 0.1 Full moon on a clear night 0.01 Quarter moon Table 3.1a Example levels of illuminance in various light conditions. Many natural scenes have fairly complex illumination, with both shadows and highlights that give different lux readings in different parts of the scene. It is important, therefore, to keep in mind that a single lux reading does not indicate the light condition for the scene as a whole. 46 Chapter 3 — Camera elements Many manufacturers specify the minimum level of illumination needed for a network camera to produce an acceptable image. While such specifications are helpful in making light sensitivity comparisons for cameras produced by the same manufacturer, it might not be so helpful when comparing cameras from different manufacturers. This is because different manufacturers use different measurement methods and have different criteria for what makes an acceptable image. To properly compare the low light performance of two different cameras, the cameras should be placed side by side and view a moving object in low light. To capture good quality images in low light or at night, Axis provides a variety of solutions such as cameras with day/night functionality, or day/night cameras with Axis Lightfinder technology or with built-in infrared (IR) LED. Cameras with day/night functionality take advantage of near- infrared light to produce quality black and white video, and day/night cameras with Lightfinder enable color video in very little light. External IR illuminators enhance the quality of black and white video in low light or complete darkness. Thermal cameras that make use of infrared radiation (at longer wavelengths than visible light) emanating from objects are the most reliable tools to detect incidents, 24/7. For more information on Axis Lightfinder technology, cameras with built-in IR LED, and thermal cameras, see chapter 2. For more information on day/night functionality and illuminators, see section 3.3. 3.2 Lenses A lens (or a lens assembly) on a camera performs several functions. > It defines the field of view, which determines how much of a scene can be seen in the image. > It preserves the details of the scene by matching the lens resolution to the sensor resolution. > It controls the amount of light reaching the image sensor so that the image is correctly exposed. > It focuses the image, either by adjusting elements within the lens assembly itself or by changing the distance between the lens assembly and the image sensor. The lens is the eye of the camera and its capabilities and features are therefore very important. Field of view, resolution, light sensitivity, and depth of field should be carefully considered and matched to your needs when choosing the camera. 3.2.1 Lens types Depending on the use there are different types of lenses to choose from: Chapter 3 — Camera elements 47 > Fixed focal length lens. Also called fixed lens. The focal length is fixed and provides a single field of view. > Varifocal lens. Offers a variable focal length and thus different fields of view. The field of view can be adjusted on the lens or through the camera’s web interface. Adjusting the focal length in a varifocal lens also requires the lens to be refocused. > Zoom lens. Is like a varifocal lens in that it offers adjustable field of view, but here there is no need to refocus if the field of view is changed. Focus is maintained when changing focal length. This lens type is very uncommon in the security industry, but the function can be mimicked by motorized lenses. Figure 3.2a Camera with a fixed lens (1), a varifocal lens (2), and a zoom lens (3). 3.2.2 Field of view The field of view describes the angle that the camera can capture. It is determined by the focal length of the lens and the size of the image sensor. The longer the focal length, the narrower the field of view. Field of view is sometimes labeled HFoV, VFoV, or DFoV, representing the horizontal, vertical, or diagonal field of view. 48 Chapter 3 — Camera elements Figure 3.2b A larger focal length (in mm) provides a narrower field of view (in degrees). A lens can be classified in one of three categories depending on which angles the lens can reproduce. > Wide angle lens. Gives a much larger field of view than what is normal for the human eye. Generally also provides a large depth of field. > Normal view lens. Gives a similar field of view as the human eye’s central field of view. > Telephoto lens. Gives a narrower field of view and provides a magnifying effect when compared to human vision. Can sometimes result in a small depth of field. Figure 3.2c Field of view with wide angle lens (1), normal view lens (2), and telephoto lens (3). The easiest way to find the required focal length for a specific field of view is to use Axis online tools Lens calculator or Product selector, both available at www.axis.com/tools. Chapter 3 — Camera elements 49 3.2.3 F-number and exposure The light gathering ability of a camera is specified by the f-number (also known as the f-stop) of the lens. The f-number defines how much light can pass through the lens and reach the image sensor. It is the ratio of the lens’s focal length to the diameter of the lens’s entrance pupil. The smaller the f-number, the better the light gathering ability, that is, more light can pass to the image sensor. In low-light situations, a lower f-number generally produces better image quality, while a higher f-number increases the depth of field. A lens with a low f-number is normally more expensive than a lens with a higher f-number. In some lenses the size of the aperture can be changed. This is done by the iris, which can be manual or controlled by the camera. When using a varifocal or zoom lens, the f-number changes when the focal length is changed. The longer the focal length, the higher the f-number. The f- number that is printed on the lens is normally valid only for the wide setting. Figure 3.2d The light gathering ability of a camera is higher when the f-number is lower. 3.2.4 Iris types The iris of a lens works in similar ways to the iris of the human eye. It controls the amount of light that passes through so that the camera image is correctly exposed. It can also be used to optimize image quality aspects, such as resolution, contrast, and depth of field. Three types of iris are common in the security industry: > In a fixed iris lens the size of the iris opening cannot be changed. This is used by the M12 (S- mount) lens. Lenses with this type of iris are mostly used in environments with controlled light levels, typically indoors. > In a DC-iris lens the camera can automatically change the size of the iris opening in response to the light level and thereby control the amount of light that reaches the image sensor. Lenses with this type of iris can be used in environments with more challenging light conditions, typically outdoors. 50 Chapter 3 — Camera elements > In a P-Iris lens the camera can control the size of the iris opening much more precisely than with a DC-iris lens. The camera can not only optimize the amount of light reaching the image sensor but also adjust for better sharpness, contrast, and a more suitable depth of field. Figure 3.2e Iris types common in the security industry: fixed (1), DC-iris (2), P-Iris (3). 3.2.5 Depth of field Depth of field refers to the distance between the closest and farthest objects that appear sharp simultaneously. This is important in applications such as the monitoring of a parking lot, where you may need to read license plates at 20, 30, and 50 meters (60, 90, and 150 feet) distance. Figure 3.2f This illustration marks out the depth of field (1) and the focal distance (2) which is the distance from the camera to its focal point. Having a larger depth of field means that objects appear sharp at a longer range around the focal point. Depth of field is affected by four factors: focal length, f-number, distance between the camera and the subject, and how the image is viewed. The part about how the image is viewed relates to Chapter 3 — Camera elements 51 aspects like the pixel size, the distance between the monitor and the observer, the observer’s eyesight, and so on. A longer focal length, a lower f-number, a shorter distance between the camera and the subject, and a shorter distance between the monitor and the observer will all decrease the depth of field. Figure 3.2g Left: photo with small depth of field — only the pens in the front are in focus. Right: photo with larger depth of field — all the pens are in acceptably sharp focus. 3.2.6 Matching lens and sensor When exchanging lenses it is important to match the lens to the camera's image sensor. If the lens is intended for a smaller sensor than the one in the camera, the image will have black corners. If the lens is intended for a larger sensor than the one in the camera, the field of view will be smaller than the lens’s capability, because part of the information outside the image sensor will be lost. This situation creates a telephoto effect as it makes everything look zoomed in. Figure 3.2h The effect of different lenses on a 1/1.8'' sensor. Right: a 1/2.7'' lens is too small for the sensor and the image has black corners. Center: a 1/1.8'' lens matches the sensor size. Left: a 1/1.2'' lens is too large for the sensor and the information outside the image sensor will be lost. The lens product pages on axis.com usually provide information about which sensor sizes are supported, and which cameras are compatible with each lens. 52 Chapter 3 — Camera elements 3.2.7 Lens types in surveillance A block lens uses motors to adjust the focus and zoom remotely as well as providing some possibilities for an optimized image quality. It is commonly used in PTZ, dome, and bullet cameras. This type of lens is built into the camera and cannot be exchanged. An M12 lens usually has a fixed focal length and no iris control. Because of its small form factor it is used in modular cameras, some dome cameras, body worn cameras, and intercoms. In some cameras this lens is exchangeable. This lens is also known as S-mount lens. A C/CS lens has a specific mounting thread, making it easy to exchange. This type of lens is used in box cameras. It exists in a variety of varifocal lengths with DC or P-Iris control. This lens offers great flexibility and is suitable for various surveillance applications. An i-CS lens has the same thread as a C/CS lens, but has extra intelligence due to built-in motors for adjusting zoom and focus remotely. It offers similar benefits as the block lens, but it is exchangeable. It is compatible with box cameras that have i-CS support. 3.2.8 Focusing Focusing a network camera often requires making fine adjustments to the lens. With the less sophisticated auto-iris lenses, it is best to adjust the focus in low-light conditions or by using a darkening filter such as a neutral density filter. In low-light conditions the iris automatically opens, which shortens the depth of field and helps the user focus more precisely. Autofocus Autofocus means that the camera automatically adjusts the lens mechanically so that the image is focused. The autofocus feature is a requirement in pan-tilt cameras where the camera direction is constantly changing. Some fixed cameras also have autofocus, but because focusing a fixed camera is normally only needed at installation, it is difficult to justify the additional cost. Chapter 3 — Camera elements 53 Some cameras include full remote focus capabilities, eliminating the need for manual adjustment on-site. The computer software gives live feedback so that the user sees that the focus is correct. With a built-in pixel counter, users can also make sure that they get enough pixels across the subject to be able to identify it, whether it is a person’s face or a license plate. The autofocus feature requires neither setting nor programming to work. In Axis PTZ cameras, it is enabled as default and starts working as soon as the camera is turned on. In scenes with low light or contrast, or that contain a lot of noise, the autofocus may need some time to find focus, and sometimes will even focus on the wrong object. When the scene changes, focus may be lost for a moment until the autofocus feature finds it again. This can give the impression that the camera is continually being focused. Focus recall A focus recall area in the desired view is a quick and easy way to regain focus immediately. The main difference between autofocus and focus recall is that autofocus will adjust focus every time the scene changes. Focus recall instead memorizes an area with a fixed focus, eliminating the need for repeated adjustments. The user sets a focus recall area by clicking the focus recall button in the camera user interface when the view has the desired focus. The camera saves its focus setting. Focus recall is then activated as soon as the user pans or tilts to the focus recall area using the joystick, and the camera automatically focuses using the saved setting. Laser focus Some lighting conditions can pose a challenge to the autofocus feature. There may be difficulties finding focus in scenes with low light or low contrast, and in scenes with reflections or pinpoint light sources. The laser focus feature makes it possible to focus on bright objects and objects that reflect a lot of light, for example window panes. Autofocus may find these challenging since the reflecting light blurs or hides the sharp edges that autofocus needs to be able to focus. With moving objects and scenes that change quickly, laser focus will find focus instantly, making it possible to focus, for example, on the license plate of a moving vehicle. 54 Chapter 3 — Camera elements Figure 3.2i Laser focus focusing on a license plate in low light conditions with pinpoint light sources Laser focus is especially useful for PTZ cameras, since the view changes dynamically when the PTZ function is used. 3.2.9 PTRZ precision Remote PTRZ (pan-tilt-roll-zoom) is a feature that makes it possible to adjust the camera view from afar, without pausing the recording. The camera can rotate around its vertical (up-and-down), lateral (side-to-side), and longitudinal (front-to-back) axes, and also change its focal length to achieve a narrower (zoom in) or wider (zoom out) field of view. 3.3 Removable IR-cut filter (day/night functionality) Many cameras have an automatically removable infrared-cut filter between the lens and the image sensor. This filter blocks the incoming infrared (IR) light, to allow the camera to produce the colors that the human eye can see. However, when the filter is removed in low light or at night, the camera sensor can then take advantage of the ambient near-infrared (NIR) light and is able to deliver black and white images in scenes with insufficient visible light. Chapter 3 — Camera elements 55 Figure 3.3a IR-cut (day/night) filter on an optical holder that, in this camera, slides sideways. The red-hued filter is used during the day and the clear part at night. 1. Solenoid 2. Front guard 3. Optical holder 4. Image sensor 5. Night filter 6. Day filter The human eye cannot see NIR light, which is in the approximate range 700–1000 nanometers (nm), but most camera sensors can detect and use it. 56 Chapter 3 — Camera elements Figure 3.3b Graph showing how an image sensor responds to visible and NIR light. 1. Relative sensor sensitivity 2. Wavelengths used in night mode 3. Wavelengths used in day mode 4. Visible light 5. Near-infrared light Cameras with a removable IR-cut filter have day/night functionality, as they deliver color video during the day, and black and white video at night. They are useful in low-light situations, covert surveillance, and in environments that restrict the use of artificial light. IR illuminators that provide NIR light can be used together with day/night cameras to further enhance the camera’s ability to produce high-quality video in low light or complete darkness. Some day/night cameras have built-in IR illumination using IR LED lights. IR illumination normally provides IR light with a wavelength of 850 nm or 940 nm. Chapter 3 — Camera elements 57 Figure 3.3c External IR illuminators and a camera with built-in IR illuminators. Built-in IR LEDs in Axis cameras can be adjusted to match the viewing angle and can be activated automatically in darkness, upon an event, or upon request by a user. Axis cameras with built-in IR LEDs simplify installation and provide a cost-effective option compared with using external IR illuminators. External illuminators, on the other hand, allow installers to freely choose the IR illuminator— for instance, a long range model—and place the light where it is needed and not necessarily in the same location as the camera. 3.4 Image sensors The image sensor is a key component of any digital camera. The image sensor registers the light coming through the lens from all parts of the scene and converts it to electric signals. These signals provide the information that is needed for the camera to, after additional amplification and processing, reproduce a digital image of the scene. The camera’s image sensor is made up of millions of photodetectors (photosensitive diodes), commonly known as pixels. Each pixel captures light (photons) throughout a defined period of time, which is the camera’s exposure time, or exposure interval. After that period of time, the pixel is read out and its charge is measured. A new exposure interval begins and the pixel can capture new photons. The quality of sensors has undergone dramatic improvements and megapixel, HDTV, and 4K sensors are widely available. But although the surveillance industry has continued to move to higher resolutions, manufacturers have often kept the same sensor size to avoid the higher cost of using a larger sensor. This means they need to fit more photodetectors in the same sensor area, making each pixel smaller and able to capture less light. The charge after each exposure interval will consequently be lower and the electric signal from each pixel will need more amplification and be noisier. Thus, by simply increasing the number of pixels in a sensor of the same size you will get better resolution, but you may also get images with lower quality. This is especially true in low light scenes, where image noise tends to be more disturbing. If you instead increase the size of the 58 Chapter 3 — Camera elements sensor, each photodetector can capture more photons and generate a stronger signal with less noise. Axis offers cameras with various sensor sizes, including several that combine 4K resolution with a large sensor. With pixels that are more than four times larger than those in most other 4K cameras, these cameras with large sensors effortlessly produce high-resolution footage that is clear and crisp even in low light. 3.5 Exposure control The two main elements that control how much light an image sensor receives are the lens’s light- gathering ability, or f-number, and the exposure time. A third element - gain - is an image level amplifier that is used to make the image brighter. However, increasing the gain also increases the level of noise (graininess) in an image, so adjusting the exposure time or iris opening is preferred. Exposure time will always have an effect on images, and the settings related to exposure can be changed in a number of ways. The most important ones, exposure priority, exposure zones, WDR and dynamic range, and backlight compensation, are explained in this section. 3.5.1 Exposure priority Bright environments require shorter exposures. Low-light conditions require longer exposures so that the image sensor can take in more light and thus improve the image quality. However, increasing the exposure time also increases motion blur and lowers the total frame rate, since more time is required to expose each image frame. In low-light conditions, Axis cameras allow users to prioritize video quality in terms of either movement or low noise (graininess). For rapid movement or when a high frame rate is required, a shorter exposure/faster shutter speed is recommended, but image quality may be reduced. When low noise is prioritized, the gain (amplification) should be kept as low as possible to improve image quality, but the frame rate may be reduced as a result. Keep in mind that in dark conditions, setting a low gain can result in a very dark image. A high gain value makes it possible to view a dark scene, but with increased noise. 3.5.2 Exposure zones A camera’s automatic exposure must decide which part of the image should determine the exposure value. In many Axis cameras, the user can also employ exposure zones to make sure that the most important part of the scene is optimally exposed. Chapter 3 — Camera elements 59 3.5.3 Backlight compensation Network cameras with backlight compensation strive to ignore limited areas of high illumination (typically the sky), as if they were not present. Backlight compensation enables objects in the foreground to be seen, although the bright areas will be overexposed. Figure 3.5a Strong backlight, without backlight compensation (left). With backlight compensation applied, the foreground is more visible (right). 3.5.4 Dynamic range Dynamic range, as it relates to light, is the ratio between the highest and lowest illumination values in an image. Many scenes have high dynamic range, with areas that are very bright and very dark. This is a problem for standard cameras, which can only handle a limited dynamic range. In such scenes, or in backlit situations where a person is in front of a bright window, a standard camera will produce an image in which objects in the dark areas will hardly be visible. To increase a camera’s dynamic range capability and enable objects in dark and light areas to be seen, various techniques can be applied. Exposure can be controlled and tone mapping can be used to increase the gain in dark areas. 60 Chapter 3 — Camera elements Figure 3.5b Two images of the same scene. The image on the right handles the dynamic range in the scene better, and details in both the bright and dark areas are visible. 3.5.5 WDR imaging Wide dynamic range (WDR) imaging is a way to recreate the full dynamic range of a WDR scene in one single image. For example, see the parking garage scene below. A conventional camera cannot capture its full dynamic range, and visibility must be sacrificed either in the dark or the bright areas. Figure 3.5c Parking garage with a challenging light situation. To the left, the image is overexposed. To the right, the image is underexposed. With a WDR-capable camera, all areas can be made visible in one single image. Chapter 3 — Camera elements 61 Figure 3.5d Parking garage with a challenging light situation, as captured by a WDR-capable surveillance camera. Axis has several solutions for WDR imaging: > Forensic WDR is a combination of a dual-exposure method and a local contrast enhancement method. It provides images that are tuned for maximal forensic usability. Employing the latest generation of image processing algorithms, this technology effectively reduces visible noise and anomalies. Forensic WDR is suitable also in scenes with motion and in ultra high resolution cameras. > WDR - forensic capture is a combination of dual exposure and a local contrast enhancement method. It provides an image that is tuned for maximal forensic usability. > WDR - dynamic capture uses a dual-exposure method for merging images with different exposure times. The dynamic range is limited by visual anomalies, for example related to motion and flickering in the scene. > WDR - dynamic contrast uses a contrast enhancement method with fairly limited dynamic range but with very few visual anomalies. Since only one exposure is used, this solution performs well in scenes with a lot of motion. The dynamic range capability of a camera is usually specified as a dB value. However, actual WDR performance is not easily measured, since it depends on many factors, including scene complexity, scene movements, and camera image processing capability. Axis prioritizes forensic usability and a high image quality instead of a high dB value. Therefore, an Axis camera with a certain specified dynamic range could very well outperform a competing camera that has a higher dB value. 3.6 Active tampering alarm The active tampering alarm automatically alerts the operator when a camera is manipulated, enabling security staff to quickly detect disrupted camera operation. The application is especially 62 Chapter 3 — Camera elements useful where there is a risk of vandalism, such as in schools, prisons, public transportation, and in harsh environments where weather, vibration, or dirt can affect the camera performance. The active tampering alarm detects incidents such as redirection, blocking, or defocusing of cameras, and reacts when the camera is attacked, spray-painted, or intentionally covered. 3.7 Audio detection Audio detection is based on the same principles as video motion detection: the application detects noise, such as voices or the breaking of a window, and uses this as a trigger to transmit and record video or audio, or to alert operators to suspicious activities. Audio detection is a powerful complement to video because it can detect activity beyond the camera’s field of view or in areas that are too dark for video motion detection. For audio detection to work, the camera needs to include audio support and either have a built-in microphone or an external microphone attached. Chapter 4 — Video encoders 63 4. Video encoders Video encoders enable an existing analog CCTV video surveillance system to be integrated with a network video system. Video encoders play a significant role in installations with many analog cameras. 4.1 What is a video encoder? Figure 4.1a Video encoders and decoders can be used to integrate analog video cameras and live view monitors in a network video system. Network cameras (1) are here supplemented by analog cameras (2) and a video encoder (3). The system is managed through a video management system (4) and can be remotely accessed from a computer or mobile device (5). A decoder (6) enables the use of monitors for displaying live video without a computer. Video encoders allow security managers to keep using their analog CCTV cameras while at the same time constructing a video surveillance system that provides the benefits of network video. If a video encoder is included in the system, analog cameras can be controlled and accessed over an IP network, such as a local area network or the internet. This also means that old video recording 64 Chapter 4 — Video encoders equipment such as DV

Technical Guide to Network Video Surveillance (PDF)

Document Details

Tags

Related

Summary

Full Transcript

Upgrade to continue