A novel neutrosophic divergence score for enhancement of mammogram images

Abstract Medical images, especially mammogram images, have low contrast and brightness and so mammogram image enhancement is very much necessary in diagnosing breast cancer or tumor at an early stage. Further, enhancement is a crucial step to increase efficiency of computer assisted hardware. This paper proposes a novel image enhancement method using neutrosophic set (NS). A variety of image enhancement algorithms are in literature, but accuracy is still a crucial problem. NS has an ability to handle indeterminant information, thus reducing the uncertainty in the images. The image is initially converted into neutrosophic domain, where an image is represented using three membership degrees – truth membership (T), indeterminacy membership (I), and false membership (F). Indeterminate degree is computed from two information, and these are combined using a novel method that uses fuzzy Lukaseiwics t norm. Then, a novel neutrosophic divergence score (NDS) is suggested, which is computed from fuzzy divergence that measures the degree with respect to an ideal image and the image so formed has better contrast with noticeable fine structures. Then a modified histogram hyperbolization is used that uses a logarithmic function to obtain a final enhanced image. Performance of the proposed method is evaluated and compared both qualitatively and quantitatively with recent methods. Experiment has been performed on different types of mammogram images to evaluate the performance of the proposed method.