Influence of the structure and parameters of CNN model on its classification ability

Convolutional neural network (CNN) is a widely used deep neural network model. It introduces convolution operation and sampling operation on the basis of artificial neural network, which greatly improves the ability of extracting signal features.

Compared with the traditional neural network, CNN greatly reduces the parameters of the model and the network structure is simpler. CNN can be regarded as a special deep neural network, and its biggest feature is the addition of convolution layer and pool layer in the hidden layer between input and output layers. This reduces the interconnection of neurons between layers and reduces the order of magnitude of calculation parameters, making CNN more suitable for image classification tasks than traditional neural networks.