Vehicle detection method based on convolutional neural network

Abstract

The invention discloses a vehicle detection method based on a convolutional neural network. The method includes the step S1 of collecting vehicle samples and non-vehicle samples and classifying the vehicle samples, the step S2 of preprocessing the samples, the step S3 of training a CNN vehicle detector, the step S4 calculating an average similarity table of a characteristic pattern, the step S5 of constructing a similarity characteristic pattern set, the step S6 of obtaining a CNN-OP vehicle detector, the step S7 of obtaining detection images, the step S8 of preprocessing the obtained detection images, the step S9 of constructing an image pyramid for the detection images, the step S10 of extracting characteristics, the step S11 of scanning characteristic patterns, the step S12 of classifying the characteristics, and the step S13 of combining detection windows and conducting output. An offline optimization scheme is put forward, the convolutional neural network which is completely trained is optimized, the strategy of scanning the windows after extracting the characteristics is adopted at the detection stage, and therefore the characteristics are prevented from being repeatedly calculated, and the detection speed of the system is increased.

Description

technical field [0001] The invention relates to a vehicle detection method based on a convolutional neural network, which belongs to the field of computer vision. Background technique [0002] However, in recent years, the number of cars has increased faster than the progress of urban road construction, resulting in urban traffic congestion and people's travel is inconvenient. In order to solve the huge pressure of urban traffic, Intelligent Transportation System (ITS) came into being. The intelligent transportation system calculates the traffic flow by detecting vehicles in different directions at the intersection, and automatically adjusts the time of traffic lights based on this, effectively improving the traffic capacity of the intersection and alleviating urban traffic congestion. Among them, vehicle detection technology is a key part of the intelligent transportation system, and subsequent more detailed analysis of vehicles must be based on accurate vehicle detection ...

AI Technical Summary

Problems solved by technology

However, all samples of this method are normalized to 32*32, which is only for the detection of front or rear vehicle detection, and the vehicle detection effect of other perspectives cannot be guaranteed, and the detection method of the scanning window is used for detection. When the detection image is For high-definition images, the detection speed cannot be guaranteed

[0006] According to the analysis of the above two traditional vehicle detection methods, there are certain defects in the vehicle detection methods based on image processing and manually designed features.

Images