Image classification method based on structure optimization sparse convolutional neural network and medium

Abstract

The invention provides an image classification method based on a structure optimization sparse convolutional neural network and a medium. For a connection structure of a convolutional layer of the convolutional neural network to an input feature map channel, a genetic algorithm is used to sparsify the convolution layer, and the sparsified convolution model I used to carry out image classification. Firstly, a convolution model is pre-trained and a pre-training weight is stored; then, according to a certain convolution layer except the model input layer, binary coding is carried out on connection of an input characteristic channel, and a plurality of binary sequences are generated and serve as an initial population; secondly, selecting, crossing and mutating of binary codes are realized byusing a genetic algorithm; and finally, after a plurality of iterations, the obtained optimal binary sequence is decoded to obtain a sparse feature channel connection structure, and the classificationaccuracy of the model is recovered through weight fine adjustment.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to the technical field of model structure sparseness and image classification methods. Background technique [0002] Convolutional Neural Networks (CNNs) have played an important role in the field of image recognition, and its powerful feature extraction ability eliminates the complicated preprocessing process necessary in traditional image recognition methods. In addition, the convolutional neural network also has the characteristics of weight sharing, local receptive field and downsampling. Compared with the earlier multi-layer perceptron, the parameters of the model are greatly reduced, and the complexity of calculation is reduced. In recent years, various improved convolutional neural networks have also achieved good image classification accuracy. [0003] There are roughly two development trends in convolutional neural networks. One is that the number of layers of th...

AI Technical Summary

Problems solved by technology

The second is that the convolutional layer becomes wider and wider. Having a larger number of convolution kernels means that more features can be extracted. However, in practical applications, the increase in the number of convolution kernels not only brings a relatively large number of calculations. There are also many redundant features, which do not contribute to the classification effect of the model.

In recent years, some model pruning methods have emerged to discard redundant connections or channels, but these methods have introduced new variables or even multiple variables to measure the importance of connections or channels, and there is no unified Metrics

Images