COLOR IMAGE SEGMENTATION BASED ON MEAN SHIFT AND NORMALIZED CUTS

An approach for Image segmentation is proposed based on mean shift algorithm and normalized cuts algorithm and its application’s implementation is proposed. The normalized cuts algorithm gives good accuracy and better segmentation compared to all most of the existing methods. By using Mean Shift algorithm on the original image to partition it into sub graphs we can create image matrices with lower dimensions. The proposed algorithm first applied Mean Shift algorithm to obtain sub graphs and then applied Normalized cut. Currency denomination and detection is an application of image segmentation. It is very difficult to count different denomination notes in a bunch. This paper propose a image segmentation technique to extract paper currency denomination. The extracted ROI can be used with Pattern Recognition and Neural Networks matching technique. First we acquire the image by simple flat scanner on fix dpi with a particular size, the pixels level is set to obtain image. Some filters and segmentation algorithms are applied to extract denomination value of note. We use different pixel levels in different denomination notes. The Pattern Recognition and Neural Networks matcher technique is used to match or find currency value/denomination of paper currency. After matching the pattern the result is converted to an audio file which helps in recognition of the given Indian currency