Image Threshold Segmentation Method Based on the Gradient Adjustment and Improved Moment Preserving

Abstract

The traditional moment preserving (MP) may cause owe-segmentation, over-segmentation and ignorance of image details, so this paper introduced a image threshold segmentation method based on the gradient adjustment and improved MP. Firstly, the method got a new image by sharpening the original image, then acquired the initial threshold by MP for the new image, afterwards, obtained the final threshold by the image histogram mean improving the initial threshold, finally, segmented the sharpened image. Experimental results showed that the proposed method can not only well resolve the problem of owe-segmentation and over-segmentation, get better segmentation result, but also retain a wealth of details and advantage of the original MP. The algorithm was simple, efficient and fast.