I need a program which detects melanoma from dermascopic images using Support Vector Machine. The code should have
- Pre-processing and segmentation
- Hair removal from the input images
- Feature selection using DWT and feature extraction using GLCM
- Extracted features should include colour (HSV, luminance, chrominance) and texture (entropy, homogeneity, contrast, saturation etc)
- classifier using SVM algorithm
- Dataset more than 200 images
- The output should display whether the input image is melanoma or not.
- GUI is not necessary but I need the values of the extracted features as well as precision and accuracy to be included in the output
- neat and easy to understand documentation of the whole process. It doesn't have to be too detailed. I just need to know what the code does and why.
This project is based on the paper "A colour-based approach for melanoma skin cancer detection" This paper is attached below
I understand it is a big project and requires time and hard work. I know I have asked for a price which is not in your budget, I have experience with ML and I can do this project. Let me know if you can afford this.