Color Image Enhancement with Different Image Segmentation Techniques

Image can be improved using different image enhancement techniques. It includes different types of operations like image segmentation, clustering, smoothing, etc. Basically in the process of Image segmentation features having homogenous characteristics are identified. The Input for the system is color image. The image gets converted into horizontal and vertical shape histogram. Then cluster formation is done using hill climbing technique and k means clustering. K Means clustering consider colour intensity as criteria. Sequential probability ratio test checks similar characteristics between different regions. Merging of these regions follows dynamic region merging algorithm. Depending on similar properties partitions are merged. The output is enhanced segmented image. Nearest neighbour graph technique is helpful to speed up the above process. This improved image is useful in the field of medical as well as security purpose.