Cross-domain image classification method based on class consistency structured learning and related device

Abstract

The embodiment of the invention discloses a cross-domain image classification method and device based on class consistency structured learning, electronic equipment and a computer readable storage medium. The method comprises the steps of: obtaining an initialized pseudo label of a target domain sample based on a first image classifier trained by using source domain data; obtaining an initialized projection matrix, weight matrix, class weight matrix, Laplacian matrix of a source domain and Laplacian matrix of a target domain based on the initialized pseudo label and a label of the source domain; updating a projection matrix by using the initialized matrixes; carrying out projection learning by using the updated projection matrix so as to update the initialized pseudo label;and finally taking the updated pseudo label as a final classification result of a target domain sample image when the number of cycles reaches a preset number of times. Therefore, by learning the Laplacian matrixes from the source domain to the target domain in the same class, the consistency and continuity of the samples in the class are improved, so that the classification performance of the model on the samples in the target domain is improved.

Description

technical field [0001] The present application belongs to the technical field of machine learning, and in particular relates to a cross-domain image classification method, device, electronic equipment and computer-readable storage medium based on class consistency structured learning. Background technique [0002] Unsupervised domain adaptation (Domain Adaptation) refers to a machine learning method that trains a model in a labeled source domain and applies it to an unlabeled target domain. [0003] There may be differences in data distribution (marginal distribution differences and conditional distribution differences) between the source domain dataset with labels and the target domain dataset without labels, which will lead to poor performance of the model when the model trained in the source domain is applied to the target domain. Significant drops ("overfitting") may occur. In order to alleviate the data distribution difference between the source domain and the target d...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present application provides a cross-domain image classification method, device, electronic device, and computer-readable storage medium based on class consistency structured learning, which can solve the problems caused by intra-class sample consistency and continuity in existing unsupervised domain adaptation methods. The performance of the model is poor, which leads to the problem of low classification performance of the model on the target domain samples

Images