This lecture overviews Image Perception. This is very important to understand image quality issues and develop many appropriate image processing operations. It covers the following topics in detail: Human Vision Modelling, Weber ratio, Mach Phenomenon. Mathematical neuron model. HVS has band-pass frequency characteristics, Laplacian-of-Gaussian retina modelling. V1 visual cortex and Gabor functions. Spatial HVS models,… Continue reading Image Perception
Archives: Resources
Resource description
Human Visual System
This lecture overviews the Human Visual System that has many applications in understanding image perception and image quality issues. It covers the following topics in detail: Human Visual System (eyes and visual pathway, stereo vision). Retina (rods, cones, ganglion receptive fields, colour theory). Visual Pathway (optic nerves, optic chiasma, optic tract, Lateral Geniculate Nucleus, Corpus… Continue reading Human Visual System
Digital Image Formation
This lecture overviews Image Formation, which is of primary importance in ensuring image quality and enabling image processing. It covers the following topics in detail: Light reflection models. Camera structure and models, e.g., Pinhole Camera. Camera Lenses. Image Formation Models. Optical sensors, e.g., CCDs. Image scanners. Image acquisition issues: gamma correction, while balance. Image quantisation… Continue reading Digital Image Formation
Image Transforms
This lecture overviews Image Transforms that are instrumental in image filtering, compression and power spectrum estimation. It covers the following topics in detail: 2D Discrete Space Fourier Transform, 2D Discrete Fourier Transform and their properties. 2D FFT algorithms: Row-column FFT (RCFFT) algorithm, Vector-radix FFT (VRFFT) algorithm, Polynomial Transform FFT (PTFFT). 2D convolution calculation using 2D… Continue reading Image Transforms
Fast 2D Convolutions Algorithms
This lecture will overview 2D linear and cyclic convolution. Then it will present their fast execution through FFTs, resulting in algorithms having computational complexity of the order O(N^2log2N). Optimal Winograd 2D convolution algorithms will be presented having theoretically minimal number of computations. Parallel block-based 2D convolution/calculation methods will be overviewed. The use of 2D convolutions… Continue reading Fast 2D Convolutions Algorithms
2D Digital Filter Design and Implementation
This lecture overviews 2D Digital Filter Design and Implementation that has many applications in digital image filtering, computer vision (template matching) and convolutional neural networks . It covers the following topics in detail: FIR filter design, Direct 2D FIR filter implementation, Digital FIR filter implementation using FFT, Block-based convolution methods, IIR filter design and implementation.