Image processing

Edge detection and segmentation is a form of image processing where the changes in intensity at the edges are used to detect lines or curves. The edges identified by edge detection are often disconnected. They are connected to form closed region boundaries which are then segmented.  The simplest method of image segmentation is called thesholding method. In this a theshold value is used to turn a gray scale image into a binary image. Clustering methods like K-means algorithm can also be used iteratively for selecting different thresholds. Here the distance between the pixel and the cluster center is  is minimized and the cluster center is recomputed by averaging all of the pixels in the cluster. Compression based methods are used to choose the optimal segmentation based on the coding length of the data. While segmentation tries to find patterns in an image and any regularity in the image can be used to compress it. For example, the contours of an image is represented as a chain code and the smoother the boundary, the shorter the encoding. Another approach is to use histogram based methods that are very efficient because they require only one pass over the pixels. The peaks and valleys in the histograms can be used to detect clusters.