Small sample image classification method based on memory mechanism and graph neural network

Abstract

The invention discloses a small sample image classification method based on a memory mechanism and a graph neural network, which is characterized in that a small sample model is helped to perform reasoning prediction by means of learned conceptual knowledge, and specifically comprises three stages of pre-training, meta-training and meta-testing, wherein the pre-training takes the trained feature extractor and classifier as initialization weights of an encoder and a memory bank; the meta-training is characterized in that features of samples of a support set and a query set are extracted through an encoder, related information of each class is mined from a memory bank to serve as meta-knowledge, and similarity between task related nodes and the meta-knowledge is propagated through a graph neural network; and the meta-test obtains a classification result through task related nodes and meta-knowledge nodes. Compared with the prior art, the method has the advantages that a human recognition process is used for reference, a memory graph augmentation network based on information bottleneck is used, well-learned conceptual knowledge is used, the model is helped to conduct reasoning prediction, the method is simple and convenient, practicability is high, and certain application and popularization prospects are achieved.

Description

technical field [0001] The invention relates to the technical field of small-sample image classification, in particular to a small-sample image classification method based on a memory mechanism and a graph neural network. Background technique [0002] The success of deep learning stems from a large amount of labeled data, and humans only need to use a small number of samples to have good generalization ability. The gap between the two has caused people to study small sample learning. Different from traditional deep learning scenarios, the purpose of small sample learning is not to classify unknown samples, but to quickly adapt to new tasks in very limited labeled data and past knowledge. [0003] Recently, the idea of ​​combining meta-learning with episode training has achieved significant advantages in solving this problem. Intuitively, using an episode sampling strategy is a promising trend to transfer knowledge from known categories (i.e. known categories with enough tra...

AI Technical Summary

Problems solved by technology

Although meta-learning and episode training strategies have achieved remarkable results in small-sample learning, most of them ignore a key issue, that is, when each scenario (episode) comes to train, how does the knowledge learned in the past face the problem? for new tasks

[0004] In the existing technology, when faced with unknown tasks, the learned concept cannot be used for reasoning and prediction by the model

Images