Novel Approach for Salient Region Detection Improved By Dynamic Region Merging and Boundary Analysis A New way to detect salient regions

Detection of salient region detection is useful for several image processing applications such as security system for Airport, Medical imaging, Object reorganization, Criminal Investigation, Retrieval of image, filter design and Retargeting an Image. A novel approach to determine salient region of images is proposed in this paper. GMM if used for clustering the colour values in the L*a*b colour space into n classes and colour spatial variance is calculated. Then dynamic region merging segmentation algorithm is utilized to extract the high level information and to achieve full resolution saliency maps. Contextaware technique is used, which makes salient regions more brilliant in the saliency map. The segmented results of image are used to compute the saliency value based on the colour contrast computation and define the saliency for each region as the weighted sum of region contrast to all regions in the image to obtain better precision and recall. The results obtained from the three modules are combined to obtain the final saliency map which is then multiplied by center distance map.