Vehicle Road Scene Recognition Based on Improved Deep Learning

Abstract

In order to improve the recognition effect of autonomous vehicles on road scenes, the new proposed activation function ReLU sigmoid is proposed based on ReLU and sigmoid model solves the problem of neuron necrosis in ReLU model. By analyzing the action principle of activation function, the key points of activation function design are given, and sigmoid and ReLU are combined in the positive and negative half axes of semantic axis to optimize the road scene recognition model. Experiments on the zhengxin expressway data set show that compared with the ReLU and LReLU models, the ReLU sigmoid model effectively improves the recognition accuracy of the convolutional neural network for the road scene, from 75.12% and 67.15% to 83.70%. It is proved that the algorithm can improve the recognition performance of the deep learning model for the road scene and alleviate the phenomenon of neuron necrosis, and improve the vehicle's perception of the road environment.