A novel fractional order Sobel operator based on texture images

Contemporary edge detection methods may not get ideal results on texture images. The fractional differential operation has the advantage of enhancing texture details features nonlinearly, a fractional order Sobel operator based on texture images is proposed when improving the typical Sobel edge detection operator, denoising the input image and quoting adaptive threshold. Both theoretical and experimental results indicate that the proposed method is able to extract the features of image edges better compared with common methods. Especially, for texture images, it is better than existing integer order differential methods since it can detect more texture edges, get a better visual effect and have an inhibitory on noise. Therefore, it is an effective edge detection method.