Multi-Column Convolution Neural Network Model Based on Adaptive Enhancement

A multi column convolution neural network model is constructed by using the idea of adaptive enhancement. It is applied to the practical application of traffic signs recognition. The data are preprocessed and the convolution neural network is trained. The high performance recognition of traffic signs is realized by the convolution neural network. The experiment proves that the convolution neural network is applied to the recognition of handwritten numerals and the identification of traffic signs.