Application Of Grey System To Medical Image Preprocessing

The image filtering and edge detection have great significance for effectively improving the display quality and visual effect of an image, and for efficiently realizing the feature extraction and classification recognition. The conventional method of the image filtering and edge detection that is used for medical images is unsatisfactory. According to the characteristics of medical ultrasonic images, a method is presented for applying a GM (1,1) accumulation model and a gray correlation analysis model of the gray system theory to image processing, which can screen noises out. The edge extraction of an image is carried out by using the gray correlation analysis method, which not only can extract the edge of an image satisfactorily, but also has the function of resisting noises, thereby testifying the practicality of this method in accordance with the edge detection of an medical image. The comparative experiment shows that the method presented in this article has the superiority to the traditional one. Finally, the image recognition is discussed. Taking pancreas cancer as an example, the feature model is established. According to the extracted image features, the image recognition is done by using the gray correlation analysis method. The recognition result is in agreement with that of the qualitative analysis.