Catalog Description: This course deals with computational methods as applied to digital imagery. It focuses on image sensing and acquisition, image sampling and quantization; spatial transformation, linear and nonlinear filtering; introduction to convolutional neural networks, and GANs; applications of deep learning methods to image processing problems; image enhancement, histogram equalization, image restoration, Weiner filtering, tomography, image reconstruction from projections and partial Fourier information, Radon transform, multiresolution analysis, continuous and discrete wavelet transform and computation, subband coding, image and video compression, sparse signal approximation, dictionary techniques, image and video compression standards, and more.

Units: 3

Prerequisites: Basic knowledge of signals and systems, convolution, and Fourier Transform.

Formats:
Spring: 3 hours of lecture per week
Fall: 3 hours of lecture per week

Grading basis: letter

Final exam status: Written final exam conducted during the scheduled final exam period


Class homepage on inst.eecs