Automatic image annotation method based on semi-supervised learning

Abstract

The invention discloses an automatic image annotation method based on semi-supervised learning. The method comprises the steps that a data set is divided into a training data set, an unlabeled data set and a test set; the SIFT feature and the HOG feature of a training sample are extracted to train an LDA_SVM classifier; color and texture features are extracted to train a neural network; unlabeleddata are used to enable two classifiers to label and predict the same unlabeled sample simultaneously; according to the contribution of the classifiers to the unlabeled sample classification accuracy,the classification results of two classifiers are weighted and fused by an adaptive weighted fusion policy, so as to acquire the final prediction label probability vector of the sample; and finally two classifiers are updated by the sample with high confidence and the predictive label thereof until the preset maximum number of iterations is reached. According to the invention, the method can makefull use of the unlabeled sample to excavate the inherent law of the image feature; the number of annotation samples required for the classifier training is effectively reduced; the annotation effectis great.

Description

technical field [0001] The invention relates to the technical field of image retrieval, in particular to an automatic image labeling method based on semi-supervised learning. Background technique [0002] With the popularization of the network and digital devices, various media image data are increasing rapidly. How to organize and manage them effectively and provide users with efficient browsing and retrieval has become an extensive research issue for researchers. [0003] Image retrieval has become a very active research field since the 1970s. At present, widely used image retrieval technologies include Text-based Image Retrieval (TBIR) and Content-based Image Retrieval (Content-based Image Retrieval). -based Image Retrieval, CBIR). Due to the obvious defects of TBIR technology, especially when the number of images is very large, the workload required to manually label images is very large, and the subjectivity and inaccuracy of manual labeling are likely to cause image m...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problem that traditional image automatic labeling still requires a large number of manually labeled training samples, and the effect of automatic labeling is not ideal when the labeled sample data is small, the present invention provides an automatic image labeling method based on semi-supervised learning, which can fully Use unlabeled samples to mine the inherent laws of image features, effectively reduce the number of labeled samples required for classifier training, and obtain better labeling effects

Images