Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Footprint image retrieval method based on space-time motion and feature fusion

A feature fusion and image retrieval technology, applied in the field of image processing and degree learning, can solve the problems of low accuracy and time-consuming, and achieve the effect of preserving integrity, overcoming obvious pressure, and improving retrieval efficiency and accuracy.

Active Publication Date: 2020-09-08
ANHUI UNIVERSITY
View PDF7 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The traditional method of image retrieval of footprints in general is to rely on the experience of experts or simple comparison algorithms. These methods not only have low accuracy
It consumes a lot of time, manpower and material resources

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
  • Footprint image retrieval method based on space-time motion and feature fusion
  • Footprint image retrieval method based on space-time motion and feature fusion
  • Footprint image retrieval method based on space-time motion and feature fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] In this embodiment, a footprint image retrieval method based on spatio-temporal motion and feature fusion mainly uses convolutional neural network and convolutional long-term short-term memory network to extract the spatio-temporal features in the sequence of footprints, and increases the performance of the network model. The data set used in the present invention contains more than 3,600 pieces of footprint data. After preprocessing, there are about 36,000 pieces of single footprint image data, including more than 100 people, and each person has at least 36 pieces of footprint data images, including barefoot footprints, Different kinds of sole pattern shoe prints, and three different walking speeds, each image is tagged with personnel ID information. Such as figure 1 Shown: The whole process can be divided into the following steps:

[0055] Step 1. Take a continuous footprint image of any test object at a certain walking speed, perform preprocessing operations of pse...

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 footprint image retrieval method based on space-time motion and feature fusion. The footprint image retrieval method comprises the following steps: 1, preparing a bundled footprint image data set; 2, establishing a bundled footprint image preprocessing module; 3, establishing a preprocessing layer of multi-scale bundled footprint images and overall normalization; 4, weight initialization; 5, establishment of a spatial feature extraction module; 6, establishing a time sequence feature extraction module; and 7, training, testing and optimizing the network. According tothe method, the spatial feature information and the time sequence feature information of the bundled footprint images are extracted, and the specific feature fusion module is combined, so that richerspatial and temporal information of the bundled footprint images can be obtained. Differential feature information among different persons can be clustered. Therefore, the accurate value of bundled footprint image retrieval can be greatly improved.

Description

technical field [0001] The invention relates to the fields of image processing and degree learning, in particular to a retrieval method for footprint images based on spatio-temporal motion and feature fusion. Background technique [0002] Due to the influence of some factors such as bones and acquired living habits, the footprint image is not easy to camouflage. Compared with other traces such as palm prints and fingerprints, it is more unique and unique. In addition to scientific research significance, the research on footprint images can also be applied in commercial, security, criminal investigation and other fields. [0003] In recent years, the rapid rise of deep learning has made a new breakthrough in image retrieval of footprints, and the neural network has a strong learning ability. With the help of deep learning, the footprint images can not only reduce the manpower and material resources for analysis and data processing, but also greatly improve the efficiency and...

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): G06K9/62G06K9/40G06N3/04G06N3/08G06F16/583
CPCG06N3/049G06N3/084G06F16/583G06V10/30G06N3/047G06N3/045G06F18/2415G06F18/253G06F18/214Y02D10/00
Inventor 唐俊鹿新王年朱明樊旭晨吴洛天李双双
Owner ANHUI UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com