Traffic image multi-type vehicle detection method based on deep study

Abstract

The invention discloses a traffic image multi-type vehicle detection method based on deep study. The traffic image multi-type vehicle detection method based on deep study firstly combines a nerve network characteristic with a region generation algorithm, realizes two processes-region generation and region determination-at the same time through using a nerve network conventional layer, uses a background model to determine a motion area of a discrete series image targeting a specific scene and provides an extra reference basis to region generation, and combines with a vehicle detection result to perform upgrading and correction on the background model according to different conditions. Besides, the traffic image multi-type vehicle detection method also brings forward a network model compression scheme to reduce model parameters and calculation time, brings forward a new detection result optimization means based on grouped error calculation to replace a conventional non-maximum-value inhibition scheme, and improves overall detection accuracy.

Description

technical field [0001] The invention relates to the technical field of vehicle detection, in particular to a method for detecting multiple types of vehicles in traffic images based on deep learning. Background technique [0002] With the development of society, people's living standards have been greatly improved in all aspects of "basic clothing, food, housing and transportation", which is reflected in the improvement of road infrastructure construction in various places and the rapid increase of the total number of cars in society in terms of "travel". However, road construction is a long-term process that requires long-term accumulation to achieve results, so it often cannot keep up with the growth rate of motor vehicles in various places. The solution to the problem is to use more scientific technical means to manage road traffic, which is intelligent transportation. system. The intelligent transportation system can count the flow of vehicles on the road, identify the d...

AI Technical Summary

Problems solved by technology

[0005] In short, the existing image-based vehicle detection technology either needs to use complex features designed by hand to provide judgment basis, or needs the region generation technology based on the underlying information of the image to provide data sources, which is not enough to deal with the processing requirements of large-scale image data, so There is a need for an accurate and fast vehicle detection method for large-scale image data with multiple scenes

Images