Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Vehicle dense target detection method based on deep learning

A target detection and deep learning technology, applied in the field of vehicle dense target detection based on deep learning, can solve the problems that complex scenes cannot be detected, target characteristics are not considered, targets cannot be recognized normally, etc., to achieve automatic recognition and identification, Improve the detection ability and solve the effect of information loss

Inactive Publication Date: 2021-06-25
SOUTHWEST JIAOTONG UNIV
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In view of the above-mentioned deficiencies in the prior art, the vehicle dense target detection method based on deep learning provided by the present invention solves the problem that the existing related target detection method does not consider the target characteristics, so that the target cannot be recognized normally and cannot detect all complex scenes. target problem

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Vehicle dense target detection method based on deep learning
  • Vehicle dense target detection method based on deep learning
  • Vehicle dense target detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] The specific embodiments of the present invention are described below so that those skilled in the art can understand the present invention, but it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art, as long as various changes Within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are included in the protection list.

[0044] Such as figure 1 As shown, a vehicle dense target detection method based on deep learning includes the following steps:

[0045] S1. Construct and train a multi-scale dimensionality reduction convolutional feature extraction network;

[0046] S2. Extract the multi-scale dimensionality reduction feature map of the image to be detected through the trained multi-scale dimensionality reduction convolution feature extrac...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a vehicle dense target detection method based on deep learning. The method comprises the following steps: S1, constructing and training a multi-scale dimensionality reduction convolution feature extraction network; S2, extracting a multi-scale dimensionality reduction feature map of the to-be-detected image through a multi-scale dimensionality reduction convolution feature extraction network; S3, generating a priori knowledge anchor box based on the historical image; S4, generating all target candidate areas in the to-be-detected image; And S5, carrying out ROIpooling processing on the target candidate area to obtain a detection result with a vehicle dense target. According to the invention, the characteristic of large scale difference of the dense target is considered, and on the basis of the faster-rcnn network, the thought of multi-stage multi-resolution and multi-size dimensionality reduction convolution feature extraction and shape prior-based anchor window generation is provided, so that the detection capability of a multi-scale dense model is improved, the problem of information loss in the existing related detection method is effectively solved, and automatic identification and discrimination of dense targets are realized.

Description

technical field [0001] The invention belongs to the technical field of intelligent transportation, and in particular relates to a vehicle dense target detection method based on deep learning. Background technique [0002] With the acceleration of our country's modernization process, the construction speed of urban infrastructure cannot meet the needs of rapid economic growth at all, and crowding of people and vehicles often occurs, resulting in traffic safety, road congestion, environmental pollution, etc. The problem is becoming more and more prominent. One of the main reasons for frequent accidents is the large traffic flow on the road, crowded vehicles, and illegal mixed traffic, which are directly related to the density of vehicles. The densely distributed vehicles in the actual scene will not only directly lead to the occurrence of traffic accidents, but also affect the service capabilities of urban infrastructure. If the intelligent traffic analysis system can detect...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/32G06K9/62G08G1/01G06N3/08G06N3/04
CPCG08G1/0104G06N3/08G06V20/54G06V10/25G06N3/045G06F18/253
Inventor 吴晓张基向重阳谭舒月
Owner SOUTHWEST JIAOTONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products