Optic Disc Detection From Fundus Images By Best-Buddies Similarity

In recent years,the detection of lesions and physiological structures from fundus images has been widely studied and applied,which is of great significance for prevention diagnosis and prompt treatment of many blinding eye diseases such as cataracts and glaucoma.However,due to the increasing of people suffering from eye diseases,and the number of ophthalmologists cannot meet the needs of such a large number of patients,more and more experts and scholars in the field of computer vision and medicine pay attention to the system of automatic recognition and diagnosis for medical images.In fundus images,the optic disc is one of the most important physiological structures.The effective detection of the optic disc is a necessary step in the automatic analysis of the fundus image.In view of interference of the non-uniform light,contrast,lesions and other factors,the external characteristics of the disc are quite different,which makes it difficult to detect the optic disc from the fundus image accurately.This paper proposes a robust template matching method based on Best-Buddies similarity to perform more efficient detection of the optic disc in the fundus image.The main work of this paper includes the following three aspects:(1)Analyze the advantages and disadvantages of traditional methodsAnalyze the advantages and disadvantages of the traditional methods of the optic disc detection from fundus images.Some researchers think that the method of detecting ODs by their appearance is effective,but these methods often fail when the fundus images have lesions and the lesions are similar to the brightness of ODs,and these methods may have high computational complexity.Vessel-based methods require a strict geometric template,and the formulation of the template depends on accurate image segmentation,which leads to a highly complex algorithm and an unsatisfied real-time requirement of OD detection.(2)Propose a method based on the Best-buddies similarity to detect optic discThere are few methods based on template matching,and some template matching methods with appropriate similarity measures show good performance in natural images.This idea is applied to medical images,experiments are performed to verify the theory.Combining the features and the location of optic disc in the fundus image,two innovation points of this paper are proposed.(1)We propose a new method for the automated detection of ODs from fundus photography.It is carried out by applying the Best-Buddies Similarity(BBS)measure based on template matching to address the problem of OD localization.(2).For capturing the spatial distribution of patches in the templates and fundus images,we adjusted the traditional BBS model byintroducing a gradient constraint term to improve the accuracy of distance measurements.(3)Design experiments to achieve our method,and make an analysis of the experimental resultsThe performance of the proposed method is validated with Digital Retinal Images for Vessel Extraction(DRIVE)and Standard Diabetic Retinopathy Database Calibration Level 1(DIARETDB1)databases,and quantitative results were obtained.Success rates/error distances of 100%/10.4 pixel and of 97.7%/12.9 pixel,respectively,were achieved.The algorithm has been tested and compared with other commonly used methods,and the results show that the proposed method shows superior performance.