digital-image-processing-hw
digital-image-processing-hw copied to clipboard
Homeworks(codes & reports) from SYSU's Digital Image Processing course taught by Professor Hongyang Chao.
digital-image-processing-hw
Homeworks(codes & reports) from 2016 SYSU's Digital Image Processing course taught by Professor Hongyang Chao.
Welcome
Welcome to my DIP homeworks! They are basic exercises for understanding DIP fundamental knowledge. Our text book is Digital Image Processing by Gonzalez and Woods.
Warning: homework solutions for reference but not Plagiarism.
If you find bugs, false logic, just anything that could be better, please do me a favor by creating issues. I would love to see constructively negative feedbacks.
And sorry that the implementation of Fast Fourier Transform in hw3 maybe wrong.
Structure
There are four homeworks and a final project in total. Each homework is composed of several questions and certain programming works. And the final project is about haze removal.
hw1-Scaling and Quantization
- Exercises
- Storage
- Adjacency
- Logical Operations
- Programming Tasks
- Scaling
- Quantization
hw2-Histogram and Spatial Filtering
- Exercises
- Linearity
- Spatial Filtering
- Logical Operations
- Programming Tasks
- Histogram Equalization
- Spatial Filtering
hw3-Filtering in the Frequency Domain
- Exercises
- Rotation
- Fourier Spectrum
- Lowpass and Highpass
- Programming Tasks
- Fourier Transform
- Bonus: Fast Fourier Transform
- Filtering in the Frequency Domain
hw4-Image Restoration and Color Image Processing
- Exercises
- Color Spaces
- Fourier Spectrum
- Lowpass and Highpass
- Programming Tasks
- Image Filtering
- Image Denoising
- Histogram Equalization on Color Images
Final Project: Haze Removal
-
Description & Background
In ECCV 2010 (one of the top conferences in Computer Vision), Kaiming He (currently a researcher in Facebook) designed a simple but powerful approach to perform as an edge-preserving smoothing operator, called Guided Image Filer. This operator is proposed as a smoothing operator, but moreover, it is effective and efficient in a great variety of computer vision and graphics applications, including details enhancement, HDR compression, image matting/feathering, etc. This approach has been included in Matlab 2014 and OpenCV 3.0 as a basic function. Considering its easy implementation, we can reproduce it in the final project.
-
Tasks
- Implement the approach described in "guided_image_filtering_eccv2010.pdf"(2013 version is better) paper.
- Evaluate the implementation by conducting experiments.
- Explore the disadvantages of the paper by experiments, and try to handle those issues.
- Reference Guided Image Filtering