Image segmentation and extraction using improved visual attention model
Ankita Panigrahi, Sunita Sarangi, Shubhendu Kumar Sarangi · 2013
Image segmentation is the process of assigning a label to every pixel in an image such that pixels with the same label share certain visual characteristics. In this paper image segmentation and extraction is described which is done using improved visual attention model. Input taken is of color image then it is converted into gray image. Gray and edge features are extracted using canny edge detector and Gabor filter respectively. After that center-surround difference operator is used to get gray and edge feature map. To get the saliency map gray and edge feature map are linearly combined. On that part image segmentation and extraction method is applied to extract each and every parts of the image. Experimental results show that this method is well organized, systematic and economical in such a way that it achieves the salient part of image which is the most informative part as well as extraction of other parts of image. So as a whole it is one of the models which automatically find Region of Interest (ROI) as well as other parts of image which is virtuous and dynamic in nature. The proposed method is effective to reduce over segmentation in auto extraction of ROI for different images. Index Terms— Improved visual attention model; Gray and edge feature extraction; Center-surround difference operator ;saliency region; Gabor filter; LOG; Canny edge detector.