APECE-505 Intelligent System Engineering Basics of Digital Image Processing! Md. Atiqur Rahman Ahad Reference books: – Digital Image Processing, Gonzalez & Woods. - Digital Image Processing, M. Joshi - Computer Vision – a modern approach, Forsyth & Ponce
Syllabus: Expert system Neural networks Fuzzy logic Robot vision – Intro, 2-stages of robot vision, image processing, genetic/pattern discovery program, scene analysis, interpreting line & curves in the image, model-based vision Genetic Algorithm
Computer / Robot / Machine vision • vs. • Human vision • Machine vs. Human • Camera vs. Eye • Computer/Processor vs. Brain • Artificial intelligence vs. Human brain… • - Very difficult for a machine – as object varies, number of object varies, dimensional issues, view-/illumination-/angle-/perspective-invariance, etc.
Computer vision • Endowing machines with the means to “see” • Create an image of a scene and extract features • Very difficult problem for machines • Several different scenes can produce identical images. • Images can be noisy . • Cannot directly ‘invert’ the image to reconstruct the scene.
CV - creates a model of the real world from images • recovers useful information about a scene from its two dimensional projections • Finding out objects in the scene • Looking for “edges” in the image • Edge: a part of the image across which the image intensity or some other property of the image changes abruptly. • Attempting to segment the image into regions. • Region: a part of the image in which the image intensity or some other property of the image changes only gradually.
Image processing stage – transform the original image into something that can be helpful for scene analysis • Interpreting lines edge detection, edge accumulation, end-point identification • Curves analysis junctions • 2. Scene Analysis stage – attempt to create an iconic [build a model] or a feature-based description of the original scene, providing a task-specific information
Robot-player • Identify lines, corners • Identify the ball [ellipse or circle] • Identify players – opponents!
MACHINE VISION Imaging device Scene Image Description Illumination Application feedback A typical CV-based control system
Machine Vision Stages Analog to digital conversion Image Acquisition Remove noise, improve contrast… Image Processing Find regions (objects) in the image Image Segmentation Take measurements of objects/relationships Image Analysis Match the description with similar description of known objects (models) Pattern Recognition
Model-based vision: • Considering various models and fit into it. • Cylindrical, stick model, etc. • e.g., Hierarchical representation through smaller cylinders to recreate a person
Stereo vision & depth information: • Stereo vision has two or more cameras • Depth info from a single camera is difficult or almost impossible – though through texture analysis, it might be possible a bit • Depth calculate the distance of foreground objects – far or closer! • Stereo vision – key constraint is correspondence problem or registration problem