Pattern recognition method and device based on convolutional neural network and computer equipment

Abstract

The invention provides a pattern recognition method and device based on a convolutional neural network and computer equipment. The pattern recognition method comprises the following steps: determininga plurality of initial neural units; taking the plurality of initial neural units as basic genes of a genetic algorithm to perform iterative computation, wherein genetic selection is performed according to the accuracy of pattern recognition of chromosome individuals in an iterative process, and the higher the accuracy of pattern recognition of the chromosome individuals is, the lower the probability that the chromosome individuals are mutated and crossed; when the iteration result meets a preset termination condition, constructing a target convolutional neural network according to the chromosome individuals of which the accuracy meets a first preset threshold; and inputting the to-be-identified object into the target convolutional neural network to obtain a mode identification result corresponding to the to-be-identified object, wherein the pattern recognition result can be stored in a block chain. According to the pattern recognition method and device, the workload of workers in theprocess of constructing the convolutional neural network model can be reduced, and the efficiency of pattern recognition is improved.

Description

technical field [0001] The present invention relates to the technical field of artificial intelligence, in particular to a convolutional neural network-based pattern recognition method, device and computer equipment. Background technique [0002] As an important model and algorithm of artificial intelligence, convolutional neural network has been widely applied to many scenarios. Especially in pattern recognition, including image understanding, natural language processing, data classification and regression, etc., all have achieved good results. Pattern recognition based on convolutional neural networks has largely replaced human work. Complex convolution Neural networks have even outperformed humans in the field of pattern recognition. [0003] However, the inventors have found that in the process of pattern recognition based on convolutional neural networks, it is first necessary to construct a convolutional neural network model, then train the convolutional neural networ...

AI Technical Summary

Problems solved by technology

[0003] However, the inventors have found that in the process of pattern recognition based on convolutional neural networks, it is first necessary to construct a convolutional neural network model, then train the convolutional neural network model, and finally use the trained convolutional neural network to perform pattern recognition. Recognition, among them, when constructing the convolutional neural network model, it is necessary to manually design the structure of the convolutional neural network, and the manual design of the structure of the convolutional neural network requires a lot of human energy and wisdom, and requires continuous testing and optimization. large amount

Images