Skip to content

Coursework for EE569 (Digital Image Processing) at USC for the Spring 2021 Semester

Notifications You must be signed in to change notification settings

ArminBaz/USC-EE569

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

51 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

USC-EE569 Spring 2021

As one can guess, this is a repo where I will store all of the homework assignments for this course (EE-569 Digital Image Processing).

The file structure should be relatively easy to follow and I have a README for each homework assignment that gives a general idea of how the folder itself is structured and how to run the code.

The general breakdown of each assignment is as follows:

Homework 1

Problem 1: Image Demosaicing and Histogram Manipulation

  1. Bilinear Demosaicing
  2. Histogram Manipulation

Problem 2: Image Denoising

  1. Basic Denoising Methods
  2. Bilateral Filtering
  3. Non-Local Means (NLM) Filtering
  4. Mixed noises in color image

Problem 3: Special Effect Image Filters: Creating Oil Painting Effect

Homework 2

Problem 1: Edge Detection

  1. Sobel Edge Detector
  2. Canny Edge Detector
  3. Structured Edge
  4. Performance Evaluation

Problem 2: Digital Half-toning

  1. Dithering
  2. Error Diffusion

Problem 3: Color Half-toning with Error Diffusion

Homework 3

Problem 1: Geometric Image Modification

Problem 2: Homographic Transformation and Image Stitching

Problem 3: Morphological Processing

  1. Basic morphological process implementation
  2. Solution to the maze
  3. Defect detection and count

Homework 4

Problem 1: Texture Analysis

  1. Texture Classification - Feature Extraction
  2. Advanced Texture Classification - Classifier Explore

Problem 2: Texture Segmentation

  1. Basic Texture Segmentation
  2. Advanced Texture Segmentation

Problem 3: SIFT and Image Matching

  1. Saliet Point Descriptor
  2. Image Matching
  3. Bag of Words

Homework 5

Problem 1: CNN Training on LeNet-5

  1. CNN Architecture
  2. Compare classification performance on different datasets
  3. Apply traed network to negative images

Homework 6

Problem 1: Origin of Green Learning

  1. Feedforward-designed Convolutional Neural Networks (FF-CNNs)
  2. PxelHop and PixelHop++

Problem 2: MNIST & Fashion-MNIST Classification

  1. Building the PixelHop++ Model
  2. Comparison between PixelHop and PixelHop++
  3. Error Analysis