A retinal image segmentation algorithm based on a residual error U-NET network

Abstract

The invention discloses a retinal image segmentation algorithm based on a residual error U-NET network, which comprises the following steps: A, downloading a color fundus retina image, and carrying out sample expansion on the downloaded image; B, preprocessing the original image; C, obtaining an Improved residual U-NET network by adding a residual structure to the U-NET network; D, taking the processed retina image as input, and carrying out pre-training on the training sample to obtain Initial parameters of the residual U-NET network model; E, Segmenting test samples using a trained residualU-NET neural network model to obtain a final retinal vessel image segmentation map; the residual U-NET neural network model iss adopted to segment the diabetic retina image, the processing effect isgood, the method can be widely applied to the field of diabetic retina diagnosis, and powerful theoretical and technical support is provided for pathological diagnosis of the diabetic retina image.

Description

technical field [0001] The invention relates to the technical field of retinal image processing, in particular to a diabetic retinal image segmentation algorithm based on a residual U-NET network. Background technique [0002] In clinical practice, color fundus images taken by ophthalmoscope are often used as the basis for screening, diagnosis and analysis of related diseases. Therefore, blood vessel segmentation of fundus images has become a prerequisite for quantitative analysis of diseases. For fundus image blood vessel segmentation, the manual segmentation method is very dependent on the operator's experience and technology, and often has shortcomings such as strong subjectivity, low repeatability, high labor intensity, and low efficiency. It is very important in application. Especially with the development of computer-aided diagnosis systems for ophthalmic diseases, the automatic segmentation of fundus blood vessels has been recognized as a very critical and challengin...

AI Technical Summary

Problems solved by technology

In addition, due to the existence of vessel intersections, branches, and centerline reflections, it is also difficult to accurately segment blood v...

Images