Method and system for training Chinese herbal medicine disease and insect pest recognition model based on semi-supervised learning

Abstract

The invention discloses a method for training a Chinese herbal medicine disease and insect pest recognition model based on semi-supervised learning, and the method comprises the steps: obtaining a labeled image data set comprising a plurality of sample images, wherein each sample image is labeled with a corresponding sample pathological type label; training a Resnet50 deep learning model based onthe sample image to obtain a first annotation model; acquiring an unlabeled image data set containing a plurality of unlabeled images, and inputting the unlabeled images into the first labeling modelto obtain pathological types corresponding to the unlabeled images and probability values corresponding to the pathological types; training a finetune model corresponding to the first labeling model based on the labeling image data set, the pathological type of each unlabeled image and the corresponding probability value so as to obtain a Chinese herbal medicine disease and pest identification model; and recognizing the to-be-labeled Chinese herbal medicine image through the Chinese herbal medicine disease and pest recognition model. The recognition accuracy is improved. The invention relatesto a smart medical scene, so as to promote the construction of a smart city.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of artificial intelligence, and in particular to a method and system for training a Chinese herbal medicine pest identification model based on semi-supervised learning. Background technique [0002] At present, deep learning is an algorithmic weapon in the era of big data, and has become a research hotspot in recent years. Compared with traditional artificial intelligence algorithms, deep learning technology has two advantages. First, deep learning technology can continuously improve its performance with the increase of data scale, while traditional artificial intelligence algorithms (including rule-based expert systems) are difficult to continuously improve their performance by using massive data. The second is that deep learning technology can directly extract features from data, reducing the work of designing feature extractors for each problem, while traditional artificial inte...

AI Technical Summary

Problems solved by technology

[0003] In terms of image data labeling of Chinese herbal medicine diseases and insect pests, the work of labeling Chinese herbal medicine diseases and insect pests image data is time-consuming, labor-intensive and expensive; Accuracy requirements

[0004] Due to the above problems, deep learning technology currently lacks a mature and effective system for image data labeling of Chinese herbal medicine diseases and insect pests, and multi-source and irregular image data labeling of Chinese herbal medicine diseases and insect pests has not been effectively utilized.

Images