In this project, we tried to implement multiple computer vision & image processing techniques, starting from the basic level of image manipulation to some advanced images segementation and texture analysis.
- Clone the repository on your local machine
- Create a virtual environemnt and activate it:
python3.8 -m venv venv
. venv/bin/activate
- Install the dependencies using
pip install -r requirements.txt
In this project, the code is structured into 4 main modules: fundamentals
, frequency-filters-segmentation
, image-segmentation
, and texture-analysis
.
In this modules, we made some basic image manipulation like:
- Image loading
- Handling channels
- Converting between channels
- Thresholding
- Histogram & intensity transformations:
- Negative transformation
- Log transformation
- Power-law transformation (gamma transformation)
- Histogram equialization
- Calculating image contrast
In this module, we did image segmentation using some frequency filtering techniques:
- Filter images using Fourier Transform
- Filter images using Wavelet Transform (also for image compression)
- Correct image using erosion and dilation
In this module, we implemented different image segmentation techniques, such as:
- Contour approach
- Detecting edges using Sobel
- Region based segmentation
- Thresholding/Otsu
- Split-merge
- Erosion & Dilation
In this module we tried some texture analysis techniques like:
- Cooccurence matrix
- LBP (Local Binary Patterns)
If you find it useful, please consider giving it a star.