Authors : D. Jeyakumari and D. Somasundareswari
Abstract: As, there is a need of image segmentation mostly in every fields like Industrial inspection, optical character recognition and all other computer related applications. It is essential step in image processing in computer vision but also a challenging procedure. In this study, introduced blend color image segmentation algorithm to overcome varies difficulties in region based segmentation. The boundaries are located by color gradient recognition method before that RGB is converted to CIE L*a*b color space using this technique histogram is formed. The histogram then generates gradient thresholds. The gradient values are increased until region is detected in different sequence. The regions with similar characteristics are merged by using statistical region merging. This process provides a very efficient procedure for an organized image segmentation.
D. Jeyakumari and D. Somasundareswari, 2016. Blend Color Image Segmentation of Statistical Region Merging Algorithm. Asian Journal of Information Technology, 15: 765-771.