High and low frequency interleaved edge feature enhancement method suitable for pedestrian target detection and method for constructing enhancement network

An edge feature and enhanced network technology, applied in the field of deep learning, can solve problems such as ignoring spatial frequency feature information, not having the ability to learn spatial features, and improving detection accuracy

Pending Publication Date: 2020-12-22
DALIAN NATIONALITIES UNIVERSITY
View PDF1 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The above two patents only process frequency features through wavelet transform, do not have the ability to learn spatial features, emphasize the transformation of the original frequency, ignore the corresponding spatial frequency feature information, and there is still room for improvement in detection accuracy

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
  • High and low frequency interleaved edge feature enhancement method suitable for pedestrian target detection and method for constructing enhancement network
  • High and low frequency interleaved edge feature enhancement method suitable for pedestrian target detection and method for constructing enhancement network
  • High and low frequency interleaved edge feature enhancement method suitable for pedestrian target detection and method for constructing enhancement network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0113] Effective combination with CornerNet algorithm

[0114] image 3 This is a visual comparison chart of feature enhancement based on the CornerNet network. In the CornerNet network, a high-low frequency interleaved edge feature enhancement network is added, and the feature edge is enhanced through the high and low frequency interleaved edge feature enhancement network, and the appropriate high-frequency feature information is fused and output. The calculation of the heat map in the next step in CornerNet provides good preprocessing. Further improve the detection accuracy, improve the false detection or missed detection of CornerNet, and improve the detection of small targets.

Embodiment 2

[0116] Vehicle Pedestrian Recognition in Traffic Roads

[0117] This example is based on Example 1. Vehicles and pedestrians in road traffic can be effectively classified, and vehicles and pedestrians in road traffic are selected as targets for detection. Figure 4 The detection results of the algorithm before and after adding high and low frequency interleaved edge features to enhance the network are given in . In the construction of intelligent transportation in the future, the target detection of pedestrians and vehicles must be the most important part, and in the test results, it can be seen that pedestrians and vehicles are effectively distinguished, and the high and low frequency interleaving edge feature enhancement network can be applied Vehicle pedestrian recognition in traffic roads.

Embodiment 3

[0119] Vehicle and pedestrian recognition in crowded roads

[0120] This example is based on Example 1. Vehicles and pedestrians in densely populated roads can be effectively classified, and vehicles and pedestrians in the road are selected as targets for detection. Figure 5 The detection results of the algorithm before and after adding high and low frequency interleaved edge features to enhance the network are given in . In dense crowds, this method can effectively detect small distant targets, effectively distinguish pedestrians and vehicles, and correctly detect all targets in the scene.

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 provides a high and low frequency interleaved edge feature enhancement method suitable for pedestrian target detection and a method for constructing an enhancement network, and belongs to the technical field of target detection. The method is characterized by comprising the following steps: S1, selecting a convolution module to perform dimension transformation, adjusting the scale ofa feature map, and extracting high and low frequency feature components according to a frequency distribution coefficient; S2, fusing the output high-frequency component with the low-frequency component through a pooling and convolution module; S3, fusing the output low-frequency component with the high-frequency component through a convolution and up-sampling module; and S4, returning the outputhigh-frequency and low-frequency fusion components to the original feature scale through deconvolution, and outputting feature fusion information under the combined action. The method has the advantages that the method can serve as an independent unit to be embedded into a deep neural network pedestrian target detection system, edge contour feature information of pedestrian targets can be remarkably enhanced, and detection precision is improved.

Description

technical field [0001] The invention belongs to the field of deep learning and relates to a pedestrian target detection edge feature enhancement network. Background technique [0002] Target detection technology is an important research field in computer vision. The target detection algorithm based on deep learning is committed to continuously improving the detection accuracy. It has a wide range of applications in the fields of smart cars, smart transportation, video surveillance, robots and advanced human-computer interaction application prospects. [0003] The existing algorithms for the fusion of image frequency information can be divided into two categories. The first category is based on the spatial domain method, which usually constructs a fusion image from the original image in the spatial domain, and performs poorly in fusing color and texture images. The other is the method based on the transform domain, which is usually divided into three steps: image decompositi...

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): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V40/10G06V10/44G06V2201/07G06N3/045G06F18/253
Inventor 杨大伟李雪萌毛琳
Owner DALIAN NATIONALITIES UNIVERSITY
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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap