Mcaspp Neural Network Fundus Image Cup and Disc Segmentation Model Based on Attention Mechanism

Abstract

Examples of the present invention involve a McASPP neural network eye -looking cup visual disk segmentation model based on the Attention mechanism.Among them, this model includes: feature extraction module, attention mapping module, multi -scale empty convolutional module, and output module. The first image feature of the input image is extracted through the feature extraction module.The second image features, and the first feature of the high -level characteristics, low -level characteristics, and the second image characteristics in the first image characteristics. Through multi -scale empty convolutional modules, the second features are obtained by multiple convolution of high -level features. The output module is output module.According to the first and second features, the predictive probability chart can improve the characteristics of the characteristics of the image segmentation network and avoid technical problems with lower accuracy of the visual cup with full convolutional network segmentation in related technologies.

Description

technical field [0001] The invention relates to the field of neural networks, in particular to a multi-scale hole convolution MCASPP neural network fundus image cup and disc segmentation model based on an attention mapping mechanism. Background technique [0002] The analysis of retinal fundus images is very important for ophthalmologists to deal with fundus diseases such as diabetic retinopathy and glaucoma, as well as other diseases related to fundus performance, such as hypertension and coronary heart disease. If not diagnosed and treated in time, there is a risk of blindness or worse. Optic cup and optic disc are one of the most basic organizational structures in retinal fundus images, and changes in the shape of the optic cup and optic disc are an important basis for clinical diagnosis of glaucoma. And in the practice of clinical diagnosis, there is a serious shortage of doctors for glaucoma, making diagnosis difficult. Therefore, automatic retinal fundus image cup-di...

AI Technical Summary

Problems solved by technology

[0006] In the traditional unsupervised method mentioned above, the neural network model usually involves many additional conditions that need to be met, and has high requirements on the quality of the image itself, and the accuracy of the segmented optic cup and disc is also low.

For the fully convolutional neural network mentioned in the neural network segmentation method, a lot of useful information is lost by extracting features layer by layer, resulting in the parameters finally learned by the model cannot fully describe the characteristics of the cup and plate.

[0007] For the above problems, no effective solution has been proposed

Images