Image Segmentation Using Distance Transform and Watershed Algorithm

This project presents an implementation of image segmentation using MATLAB software. Image segmentation refers to the process of partitioning a digital image into multiple segments or regions, where each region represents a meaningful component of the image. We employ a combination of distance transform and watershed algorithm to achieve effective segmentation. The distance transform method calculates the Euclidean distance from each pixel to the nearest boundary pixel, which can be implemented using MATLAB's bwdist() function. The watershed algorithm, implemented through watershed() function, treats the image as a topographic surface and floods basins from markers to separate different regions. Our implementation involves key steps: preprocessing with morphological operations, distance transform computation, marker extraction using regional maxima, and watershed transformation. This combined approach enhances segmentation accuracy by preventing over-segmentation through proper marker selection and improves computational efficiency through optimized distance mapping. The methodology successfully achieves precise image segmentation by effectively distinguishing between adjacent objects and background regions.