Lesion detection method and apparatus of eye fundus image

Abstract

The application discloses a lesion detection method and apparatus of an eye fundus image. The method comprises the following steps: carrying out pretreatment on a to-be-detected eye fundus image; carrying out optic disk positioning on the eye fundus image after pretreatment and determining an optic disk region; carrying out morphological processing on the eye fundus image after pretreatment to obtain a healthy background image; and according to the healthy background image and the optic disk region, determining an effusion region in the eye fundus image after pretreatment. Therefore, the detection speed of the eye fundus image can be increased and accuracy of positioning of a lesion region can be improved.

Description

technical field [0001] The invention relates to the field of computer vision information, in particular to a method and device for detecting fundus image lesions. Background technique [0002] With the development of computer graphics processing technology, the analysis of fundus images no longer depends solely on ophthalmologists' naked eyes. As early as the 1970s and 1980s, foreign scholars had proposed the technology of automatic detection of hard exudation and hemorrhage based on fundus image processing and analysis, and conducted a lot of research. Relying on computer graphics processing technology to quickly and accurately identify lesions in fundus images can effectively solve the problem of lack of quantitative means in traditional manual qualitative analysis, and also save a lot of manpower, material resources and time, providing a great opportunity for the implementation of large-scale screening of lesions basic condition. The research results have great practica...

AI Technical Summary

Problems solved by technology

However, the above methods may have boundary fractures, making filling operations difficult, and thus missing some lesion areas

[0005] In general, the existing fundus image processing methods mainly have the following defects: low precision, unable to conduct accurate quantitative analysis; need fixed parameter combination, low universality, cannot perform accurate analysis for various images; require a large number of The data is used as a training sample; some algorithms are too expensive to calculate, and the calculation cost is too high, which exceeds the computing power of general computers

[0006] It can be seen that although there are a large number of research results on fundus image analysis, the processing speed or lesion area positioning effect needs to be improved.

Images