Colon Polyp Image Enhancement And Segmentation Algorithm Based On CA

Medical image contains abundant information about the function and structure of biomedical anatomy.It can provide support for the medical diagnosis and treatment by reading and understanding the information correctly.The good enhancement of colonic polyp images is helpful for doctors to observe the image information clearly,and further improve the accuracy of the colon diagnosis.Besides,it's the polyp type that determines whether the polyp needs to be removed,when a suspicious polyp is found in colonoscopy.The segmentation needs to be performed before the classification of polyps.And the better the performance of the segmentation,the higher the accuracy of the classification.Therefore,the enhancement and segmentation of colonic polyp images play an important role in the diagnosis of colonic diseases.Because cellular automata has the characteristics of locality,parallelism and homogeneity,and the spatial distribution structure of CA is similar to the digital storage structure of image,CA model has been widely applied in the field of image processing.The thesis studies the enhancement and segmentation algorithm of colon polyp image based on the CA model.The main contents and innovations in the thesis are as follows:1.In the aspect of colon polyp image enhancement,the thesis realize the enhancement of color colonic polyp images by applying a gray enhancement method to luminance component.For the gray enhancement method,a new image enhancement algorithm is proposed in this paper by using CA and combining the idea of the histogram equalization.The new algorithm includes two parts: global processing and local processing.In the global processing,the concept of extended histogram is proposed.Then equalize the extended histogram to weaken the impact of the traditional histogram peak;In the local processing,a CA model is established.On the basis of that the extended histogram equalization has enlarged the dynamic range of gray levels,the intensity of pixels will be mapped to a new intensity according to the CA local rule.The new algorithm combines the global and local information of the image,which can not only improve the contrast of the image,but also preserve the details of the original image.Finally,the simulation result shows that the new algorithm can achieve better results.2.In the aspect of colon polyp image segmentation,an automatic segmentation algorithm for colon polyp images is proposed by combining the Shape-UCM algorithm and the FCA model in the fuzzy Grow Cut segmentation algorithm.Firstly,guide automatic selection of the initial seed,according to the characteristics of Shape-UCM algorithm which obtaining partial or complete polyp region.Then a specific membership function is established according to the prior knowledge of the polyp shape,and added to the FCA model,to segment polyp region in colonic polyp image automatically.Finally,the simulation result shows that the proposed algorithm can obtain better segmentation results.