Zero sample image classification method based on combination of variational autocoder and adversarial network

Abstract

The invention discloses a zero sample image classification method based on combination of a variational autocoder and an adversarial network. Samples of a known category are input during model training; category mapping of samples of a training set serves as a condition for guidance; the network is subjected to back propagation of optimization parameters through five loss functions of reconstruction loss, generation loss, discrimination loss, divergence loss and classification loss; pseudo-samples of a corresponding unknown category are generated through guidance of category mapping of the unknown category; and a pseudo-sample training classifier is used for testing on the samples of the unknown category. The high-quality samples beneficial to image classification are generated through theguidance of the category mapping, so that the problem of lack of the training samples of the unknown category in a zero sample scene is solved; and zero sample learning is converted into supervised learning in traditional machine learning, so that the classification accuracy of traditional zero sample learning is improved, the classification accuracy is obviously improved in generalized zero sample learning, and an idea for efficiently generating the samples to improve the classification accuracy is provided for the zero sample learning.

Description

technical field [0001] The invention belongs to the technical field of zero-sample image classification, and in particular relates to a zero-sample image classification method based on a combination of a variational autoencoder and an adversarial network. Background technique [0002] With the rapid development of information technology, pattern recognition is an important part of information science and artificial intelligence, mainly used in image processing, speech recognition, data mining and other disciplines. The main purpose of studying pattern recognition is to classify samples. The current effective method is supervised learning, that is, training the model through a large amount of labeled data, and then testing it on the test set. But in reality, it is difficult to obtain a large number of labeled pictures, and sometimes the labeled pictures obtained are not the category pictures that need to be classified. Therefore, the study of zero-shot learning is of great s...

AI Technical Summary

Problems solved by technology

[0008] The technical problem to be solved by the present invention is to provide a zero-shot image classification method based on the combination of variational autoencoder and confrontation network, which can make up for the lack of training samples in zero-shot learning. The generated pseudo-samples transform zero-sample learning into supervised learning in traditional machine learning, thereby improving the classification accuracy rate of traditional zero-sample learning, and the classification accuracy rate in generalized zero-sample learning is also significantly improved

Images