Traditional Chinese medicine tongue picture greasiness classification method containing noisy annotation data

Abstract

The invention relates to a traditional Chinese medicine tongue picture greasy classification method with noisy annotation data, and belongs to the field of computer vision. According to the method, a noisy annotation tongue picture greasy data set is subjected to label confidence evaluation and updating processing through a plurality of machine learning classification models; label data set labeling quality updating and classification network model parameter updating are put in an iteration process, an active learning idea is introduced, and a novel traditional Chinese medicine tongue picture greasy sample labeling quality and classification model interactive iteration improvement method is provided. According to the method, the confidence coefficient of the labeled samples is improved, so that high-quality labeled samples are screened, the method can be applied to other tongue picture samples containing noise labels, the problems that tongue picture labeling in traditional Chinese medicine is high in subjectivity and contains noise labels can be effectively solved, and the method has higher generalization performance.

Description

Technical field [0001] The present invention is in the field of computer vision, particularly to Pythium Tongue tired noisy labeling method of classification data. Background technique [0002] Tongue, the tongue by observing changes, to understand the physiological functions and pathological examination methods case body, a Chinese medical broader application, one of the most popular diagnostic method. Tongue is the core content analysis features to achieve the objective of the tongue, the tongue is an inherent feature of human blood and viscera cold occur, hot, true, real pathological changes reflect. In tongue analysis usually necessary to analyze different aspects of the tongue characteristics, e.g. tongue color, thickness and moisture of tongue, tongue rot greasy, prick, crack morphology tongue (scalloped, skew and fat or thin) and other features. Rot and greasy fur texture features are reflected mainly refers to a change of the particle size of the tongue and coarseness, ge...

AI Technical Summary

Problems solved by technology

[0004] The cost of tagging the sample data of the characteristics of rancid tongue image in traditional Chinese medicine is high, and there are few samples labeled with rancidity. It is necessary to effectively use unlabeled samples to distinguish the characteristics between different categories; Therefore there are inconsistent noisy labeled samples

Images