Author(s)

Wang Jinghe

Corresponding Author

Wang Jinghe

Abstract

In this paper, the convolution neural network structure is used for image recognition. The initialized image is convoluted with the convolution core in the convolution layer to extract the features of the image. The extracted image features are compressed by the pooling layer by the full connection layer. The convolution neural network iterates the image for several times and compresses the image layer by layer to recognize the image and output the recognized image. Experiments show that compared with BP neural network, the network structure can significantly improve the accuracy of image recognition.

Keywords

Image recognition, Accuracy rate, Convolution neural network