Traditional Chinese medicine image classification and retrieval method based on pyramid network

Abstract

The invention designs a traditional Chinese medicine image classification and retrieval method based on a pyramid network, and is used for solving various problems that traditional neural network classification focuses on global semantic information such as contours and shapes, and a lot of information is lost due to multiple down-sampling operations. In order to achieve an ideal classification effect, multi-scale feature fusion is realized through a pyramid structure of a feature pyramid, and the requirement that shallow-layer and deep-layer spatial and semantic information needs to be fused in traditional Chinese medicine decoction piece image classification is met. Aiming at the problems that a multi-scale feature fusion mode can cause a large number of redundant features and neural network training can be interfered by redundant information, an SE-Net model is used, the redundant features are reduced, a sampling strategy and triple loss are combined, a multi-task learning framework is used for classifying and retrieving images at the same time, and more compact and accurate traditional Chinese medicine image feature information can be obtained.

Description

Technical field [0001] The present invention relates to a medical image processing and image classification and retrieval, the particular design of the classification and retrieval of an image based on traditional Chinese medicine pyramid network. Background technique [0002] Chinese Herbal Medicine Chinese herbal medicine is cooked according to traditional Chinese medicine, the processing processing can be directly used in clinical medicine. Receive different processing methods and slice method, the same type and size and shape of slices can not be determined. In clinical and production, its rapid classification requires a lot of professionals have the knowledge of traditional Chinese medicine. Because the main virtue of the senses and professional experience to judge, it is more time-consuming, high labor costs. Use deep learning can quickly complete classification of Chinese Herbal Medicine, traditional Chinese medicine classification of experience into a measure indicators, ...

AI Technical Summary

Problems solved by technology

[0004] The present invention designs a classification and retrieval method of traditional Chinese medicine images based on a pyramid network, which is used to solve the problem that the traditional neural network classification focuses on global semantic information such as outline and shape, and multiple down-sampling operations lead to the loss of a lot of spatial information and detailed information and other problems, in order to achieve the ideal classification effect, the image data set of traditional Chinese medicine decoction pieces is firstly cleaned and augmented, and used for classification and retrieval of the training network. Under the premise, multi-scale feature fusion is realized to meet the needs of image classification of Chinese herbal medicine pieces that need to fuse shallow and deep spatial and semantic information

Images