Retina hemorrhage identification method and system based on multi-scale residual edge detection

The embodiment of the invention provides a retina hemorrhage identification method and system based on multi-scale residual edge detection, and belongs to the technical field of image processing. The method comprises the following steps: collecting eye image information of a user; taking the eye image information as an input parameter, and carrying out pre-constructed retina hemorrhage recognition model training to obtain a training result; wherein the pre-constructed retina hemorrhage recognition model is obtained based on training of a pre-constructed MR-HED network; the pre-constructed MR-HED network comprises a residual error network structure; performing post-processing optimization on the training result to obtain an optimized result; and obtaining a retina hemorrhage identification result based on an optimized result, and pushing the identification result to a user side. According to the scheme, the image effect can be enhanced while the accuracy of an identification result is ensured. 本发明实施例提供一种基于多尺度残