Image segmentation and its application in importance map.

摘要

The objective of this study is to demonstrate the application of factors which influence the eye movements and attention and how it can be used in an algorithm to calculate the contrast of an image with respect to its background and also to detect the size of the different pixels in the image. The prime study involves the role of segmentation process. Various segmentation techniques will be discussed. The implemented algorithm involves two different segmentation techniques, the watershed technique and the quad tree decomposition method. The segmented image is then used to find the contrast factor and the size factor which helps us to understand the Human Visual System (HVS) better. The algorithm works for grey scale images and needs some improvisation for it to work for a colored image. A comparison of the outputs for both the segmentation techniques has been documented. The algorithm has been implemented using MATLAB.