Image-Segmentation-of-2D-Images-using-Clustering-based-Image-Thresholding

Image Segmentation into foreground and background using OTSU method

Steps used in OTSU Algorithm:

1. Compute the histogram of pixel intensities vs the number of pixels
2. Loop through the grayscale intensities from 0 to 255, setting each as a threshold
3. Compute the weighted mean, and the variance from the function
4. Compute the function value at that point
5. Update minimum variance and then update the threshold

---

Example implementation The input image is:-


The histogram for the given image is:-


The output of the binarization is:-

Dependencies

• Python 3
• OpenCV
• Matplotlib
• Numpy