Computer Vision and Image Processing Quiz

Computer Vision Overview

• Computer vision is a field of artificial intelligence (AI).
• It enables computers and systems to derive meaningful information from visual inputs (images, videos).
• It allows computers to "see," observe, and understand visual information.

Computer Vision vs. Computer Graphics

• Computer vision recognizes environments using images.
• Input: Images.
• Output: Description (e.g., object locations, dimensions).
• Computer graphics generates synthetic images from described information.
• Input: Description.
• Output: Images.

Computer Vision Pipeline

• Image acquisition: digital images from sensors (cameras, radar, etc.).
• Pre-processing: essential steps to prepare the image (noise reduction, contrast enhancement).
• Segmentation: partitioning the image into regions of similarity.
• Feature extraction: identifying measurable characteristics (edges, corners).
• Feature selection: prioritizing relevant features.
• High-level processing: classifying objects and estimating properties.

Computer Vision Tasks

• Object detection and recognition: identifying and classifying patterns in images.
• Face recognition: identifying and classifying human faces.
• Content-based image retrieval: finding images based on their content (color, shape).
• Optical character recognition (OCR): converting handwritten text to digital text.

Image Processing

• Image processing manipulates images to enhance quality or prepare for analysis.
• Input: Images.
• Output: Images.
• Image processing techniques are often used as steps in computer vision systems.

Filters

• Linear filters (e.g., uniform, triangular, Gaussian): reduce noise by averaging pixel values.
• Non-linear filters (e.g., median, max, min): suppress certain types of noise and preserve edges.

Coordinates and Images

• 2D coordinates: two numbers specify a location on a plane.
• 3D coordinates: three numbers specify a location in 3D space.
• Right-handed and left-handed coordinate systems are used in 3D.
• Digital images represent an image as a finite set of digital values, called picture elements or pixels.
• Color depth: the number of bits to store a color value of a pixel.

Image Categories

• Binary images: only two values (black or white).
• Grayscale images: shades of gray.
• Color images: RGB color space (combination of red, green, and blue).
• Multispectral images: capture intensities outside the visible portion of the electromagnetic spectrum.

Digital Filters

• An image can be filtered in the spatial or frequency domain.
• Spatial filtering involves convolving the image with a filter kernel.
• Convolution: a mathematical operation fundamental to many image processing operators.

Image Noise

• Noise degrades image quality (e.g., dust on lens).
• Smoothing operations can be used to reduce noise.
• Types of noise: salt and pepper, Gaussian.

Computer Vision Applications

• Manufacturing (correct product positioning).
• Visual auditing (detecting issues in equipment such as trucks, planes, etc.).
• Medical image processing (tumor detection).
• Automotive industry (object detection, parking assistance).
• Social commerce (finding similar homes based on image).
• Social listening (detecting product logos in social media).
• Retail (finding an item's price in various stores).
• Education (locating similar educational materials).
• Public safety (license plate reading).

Grading Policy

• Quizzes (practical): 10%.
• Assignments: 10%.
• Project: 10%.
• Midterm exam (theoretical): 30%.
• Final exam (theoretical): 40%.
• Total: 100%.

Test your knowledge of computer vision and image processing concepts. This quiz covers topics like distance metrics, feature matching, and the primary tasks in computer vision. Challenge yourself to distinguish between different algorithms and their applications in these fields.