Digital Image Processing Lesson 1 PDF

By Dr. Lochandaka Ranathunga 1 Overview  Digital Image Fundamentals  Image Enhancement  Edge and Boundary Detection  Image Segmentation  Image Transformations  Shapes and Features  Image Coding Compression...

By Dr. Lochandaka Ranathunga 1 Overview  Digital Image Fundamentals  Image Enhancement  Edge and Boundary Detection  Image Segmentation  Image Transformations  Shapes and Features  Image Coding Compression 40%  Image Content Analysis 2 Assignments  Image Retrievals 2 1  References  Digital Image Processing, Rafael C. Gonzalez and Richard E. Woods, Pearson Education  Image Processing : Principles and Applications, Tinlu Acharya, Ajoy K. Roy, Willey Publication  Fundamentals of Digital Image Processing: A Practical Approach with examples in Matlab 3 Introduction to Digital Images  Audio signals able to Stimulate auditory senses of human  Visual representations are often the most efficient way to represent information  Static Visual information such as pictures able to stimulate eyes  Picture can describe an incidence of a story  Video is a temporal media which stimulates both visual and auditory senses of human  Video able to describe a story with sequence (array of images) 4 2  Why are digital images interesting?  Humans are visual creatures in a visual world  Images are (often) the primary sense  So, if we want to build systems capable of human skills, then they should be capable of understanding images (many applications)  Images in a computer are DIGITAL, as opposed to analog images in an (old) photo- camera 5 Why should we learn about digital images?  Understand the visual media: Images  Understand how images can be manipulated in order to create visual effects  Rotation, scaling, blurring, etc.  As done in standard image manipulation tools  Remove parts of an image  Combine graphics with real images  Combine (part of ) one image with another  Generate control signals for an application (by forecasting)  Understand how to find and follow (well defined) an objects in an image  Recognize objects (many industrial applications) 6 3  Working with images  Image manipulation  Simple operations, e.g., scale image  Image processing  Improve the image, e.g., remove noise  Image analysis  Analyze the image, e.g., find the person in the image  Machine vision  Industry, e.g., Quality control, Robot control  Computer vision  Everything: multiple cameras, video-processing, etc. 7 Computer Vision Enable machines to “see” the visual world as we do 8 4  Fundamental Steps in Computer Vision Point 1: 22,33 Point 2: 24,39 ….. Representation Segmentation and description Actor sitting Preprocessing Recognition Knowledge base and Problem Interpretation Result domain Image acquisition 9 Computer Vision  Automatic understanding of images and video 1. Computing properties of the 3D world from visual data (measurement) 10 5  1. Vision for measurement Real-time stereo Structure from motion Tracking NASA Mars Rover Snavely et al. Demirdjian et al. Wang et al. Automatic understanding of images and video 1. Computing properties of the 3D world from visual data (measurement) credit: Kristen Grauman 11 Computer Vision  Automatic understanding of images and video 1. Computing properties of the 3D world from visual data (measurement) 2. Algorithms and representations to allow a machine to recognize objects, people, scenes, and activities (perception and interpretation) credit: Kristen Grauman 12 6  2. Vision for perception, interpretation amusement park Objects sky Activities Scenes The Wicked Cedar Point Locations Twister Text / writing Faces ride Ferris Gestures wheel Motions ride 12 E Emotions… Lake Erie water ride tree tree people waiting in line people sitting on ride umbrellas tree maxair carousel deck bench tree pedestrians credit: Kristen Grauman 13 13 Computer Vision  Automatic understanding of images and video 1. Computing properties of the 3D world from visual data (measurement) 2. Algorithms and representations to allow a machine to recognize objects, people, scenes, and activities. (perception and interpretation) 3. Algorithms to mine, search, and interact with visual data (search and organization) credit: Kristen Grauman 14 14 7  3. Visual search, organization Query Image or video Relevant archives content credit: Kristen Grauman 15 15 Related disciplines Artificial intelligence Machine Graphics learning Computer Image vision Cognitive processing science Algorithms credit: Kristen Grauman 16 16 8  Vision and graphics 2D TO 3D Images Vision Model Graphics 3D TO 2D Inverse problems: analysis and synthesis Credit: Kristen Grauman 17 17 Examples: Image Correction  Needed when image data is erroneous:  Bad transmission  Bits are missing: Salt & Pepper Noise 18 9  Image Deblurring: Motion Blur  Can be used when a camera or object is moved during exposure 19 Deblurring  Can be used when the camera was not focused properly!! 20 10 21 Image manipulation  Image improvement, e.g. too dark image  Rotate + scale 22 11  Medical Image Processing  Image Processing is widely used  E.g. Analysis of microscopic images 23 24 12  Medical imaging Image guided surgery 3D imaging Grimson et al., MIT MRI, CT Source: S. Seitz 25 Medical Imagine Global Motion K. A. S. H. Kulathilake, L. Ranathunga, G. R. Constantine, and N. A. Abdullah, “Reduction of motion disturbances in coronary cineangiograms through template matching,” in Future Information Technology, ser: Lecture Notes in Electrical Engineering. Berlin, Germany: Springer, 2014, vol. 309. 26 13  Medical Imagine K. A. S. H. Kulathilake, L. Ranathunga, G. R. Constantine, and N. A. Abdullah, “Reduction of motion disturbances in coronary cineangiograms through template matching,” in Future Information Technology, ser: Lecture Notes in Electrical Engineering. Berlin, Germany: Springer, 2014, vol. 309. 27 Autonomous agents Mars rover Google self-driving car 28 14  Autonomous Vehicles B. S. S. Rathnayake and L. Ranathunga, "Lane Detection and Prediction under Hazy Situations for Autonomous Vehicle Navigation," 2018 18th International Conference on Advances in ICT for Emerging Regions (ICTer), Colombo, Sri Lanka, 2018, pp. 99-106. R. P. U. N. Rajapaksha and L. Ranathunga, "Optimized Multi-Shaped Traffic Light Detection and Signal Recognition with Guided Framework," 2018 18th International Conference on Advances in ICT for Emerging Regions (ICTer), Colombo, Sri Lanka, 2018, pp. 1-6. 29 Conveyer belt applications  Checking and sorting  For example: checking bottles in the supermarket  Quality control  Does the object have the correct dimensions, color, shape, etc.?  Is the object broken?  Robot control  Find precise location of the object to be picked 30 15  Character Recognition Character/Word Segmentation Vertical projection profile  Line Segmentation  Horizontal projection profile “Sinhala Handwriting Recognition Mechanism Using Zone Based Feature Extraction”, Dharmapala, KAKND; Wijesooriya, WPMV; Chandrasekara, CP; Rathnapriya, UKAU; Ranathunga, L, ITRU Symposium 2015 31 Object Segmentation In Video Ground-truth Original K-Means-8D RSHC Ravimal Bandara, L. Ranathunga, Nor Aniza Abdullah 32 16  Biometrics  Recognizing/verifying the identity of a person by analyzing one or more characteristics of the human body  Characteristics:  Fingerprint, eye (retina, iris), ear, face, heat profile, shape (3D face, hand), motion (gait, writing), …  Applications:  Verifying: Access control (bio-passports)  Recognizing: Surveillance Face recognition systems Fingerprint scanners 33 Chroma keying 34 17  Analysis of Sport Motions  Here: Analysis of motion of Sarah Hughes  3D Tracking of body parts  Motion interpretation  Action recognition 35 Motion Capture  Special effects  Advertising  Movies Andy Serkis 36 18  Motion Capture 37 Braille Document Reader Scanned image of Sinhala printed Braille Text paper 38 19  Questions? 39 20