This is often the "fun" part for students. It involves improving the visual appearance of an image or converting an image to a form better suited for analysis by humans or machines. Key techniques discussed include:
Rather than just describing convolution, the best PDFs offer side-by-side comparisons of averaging vs. median filters for salt-and-pepper noise, and Sobel vs. Canny for edge detection in cluttered backgrounds. They also discuss computational complexity—vital for video. practical image and video processing using matlab pdf
Using phase correlation or feature matching ( estimateGeometricTransform ), you can remove camera shake from handheld footage—a common requirement in action cameras and drones. This is often the "fun" part for students
needing quick prototyping for medical imaging or autonomous systems. median filters for salt-and-pepper noise, and Sobel vs
It covers both classic image processing (filtering, transforms, morphology) and video processing (motion estimation, object tracking, background subtraction) – something many introductory texts ignore.