Eye fundus image blood vessel segmentation method based on linear filtering and deep learning

The invention relates to a fundus image blood vessel segmentation method based on linear filtering and deep learning, and belongs to the field of medical image processing. The method comprises the following steps: S1, inputting an eye fundus image, and enhancing a blood vessel region by using a linear filtering algorithm based on a Hessian matrix; s2, adopting MobileNetV3 as a basic model of a blood vessel segmentation model, establishing a segmentation network VSegNet, and then adding an encoder based on a recursion module into the segmentation network VSegNet to perform down-sampling; s3, adding a decoder into the segmentation network VSegNet to perform up-sampling and aggregation on the feature map output by the encoder; and S4, when the segmentation network VsegNet is trained, calculating a loss value of a segmentation result by adopting a segmentation prediction result and an L1 norm of the segmentation truth value image. According to the method, the feature information extraction capability is enhanced,