Method and device for feature matching and image recognition equipment

A feature matching and sub-pattern technology, applied in the field of image recognition, can solve the problems of prone to misidentification and time-consuming, and achieve the effects of avoiding misidentification, high matching accuracy, and improving matching efficiency

Active Publication Date: 2018-03-06
TCL CORPORATION
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Embodiments of the present invention provide a feature matching method, device, and image recognition device, aiming to solve the problem that the feature matching method provided by the prior art is time-consuming or prone to misidentification

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
  • Method and device for feature matching and image recognition equipment
  • Method and device for feature matching and image recognition equipment
  • Method and device for feature matching and image recognition equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] figure 1 The implementation flow of the feature matching method provided by Embodiment 1 of the present invention is shown, and the details are as follows:

[0067] In step S101, the features obtained at the current time point are stored in the sub-pattern library.

[0068] In this embodiment, the features in the pattern library and the features in the test library can be described by certain character strings, numbers or symbols.

[0069] In this embodiment, Lib_ti (i=1, 2, 3...N) is used to represent the features obtained at time ti. Among them, ti represents time, Lib_t1 represents the feature obtained at time t1, Lib_t2 represents the feature obtained at time t2, the larger the i in ti, the longer the time experience, and each Lib_ti can be a string of characters, numbers or symbols To represent.

[0070] Test_ti (i=1,2,3...M) represents the characteristics of the test library obtained at time ti, the larger the i in ti, the longer the time experience, and each T...

Embodiment 2

[0095] Figure 5 The implementation flow of the feature matching method provided by Embodiment 2 of the present invention is shown, and the details are as follows:

[0096] In step S501, the features obtained at the current time point are stored in the sub-pattern library.

[0097] In step S502, the sub-pattern library is added into the pattern library.

[0098] In step S503, the features of the sub-pattern library are divided into p feature segments.

[0099] In step S504, the p feature segments of the features of the sub-pattern library are correspondingly mapped to corresponding hash addresses of p hash tables through p hash functions.

[0100] In this embodiment, the features of the sub-pattern library are divided into p feature segments, such as Image 6 shown. from Image 6 As can be seen from , the sub-pattern library is characterized by characters, numbers or symbols, and these characters, numbers or symbols need to be segmented in practice, and here they are divi...

Embodiment 3

[0112] Figure 10 A specific structural block diagram of the feature matching device provided by Embodiment 3 of the present invention is shown. For convenience of description, only parts related to the embodiment of the present invention are shown. The device 10 can be built into a computer, or can be built into a special image recognition device to complete image recognition, such as fingerprint recognition. The device 10 includes: a sub-pattern library creation unit 101, a sub-pattern library storage unit 102 , a sub-pattern library mapping unit 103 , a first request receiving unit 104 , a sub-test library mapping unit 105 , a first matching unit 106 and a second matching unit 107 .

[0113] Wherein, the sub-pattern library creation unit 101 is used to store the features obtained at the current time point into the sub-pattern library;

[0114] A sub-pattern library storage unit 102, configured to add the sub-pattern library to the pattern library;

[0115] A sub-pattern l...

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 present invention is applicable to the field of image recognition, and provides a feature matching method, device and image recognition equipment. The method includes: a sub-pattern library stores features obtained at a time point, and after the sub-pattern library is added to the pattern library, It is only necessary to map the characteristics of each added sub-pattern library to the corresponding hash address of the hash table. When it is necessary to match the test library with the pattern library, first map the newly added sub-test library to the corresponding hash address of the hash table. Hash address, and then match the characteristics of the sub-test library with the characteristics in the mapped hash address to obtain the matching result. Compared with the prior art, the present invention does not need to map the entire pattern library to the corresponding hash address of the hash table every time the matching is performed, which consumes little time and is beneficial to improve the matching efficiency. Moreover, the use of the hash function for matching can avoid the problem of misidentification when using the Bron filter, and the matching accuracy is very high.

Description

technical field [0001] The invention belongs to the field of image recognition, in particular to a feature matching method, device and image recognition equipment. Background technique [0002] In the process of image feature comparison, in many cases, the features of the existing pattern library are calculated in advance, and then stored in a specified location, and finally the features in the test library and those stored in the pattern library are obtained. The features are compared to obtain the comparison result. The pattern libraries mentioned here are often completed or established offline in advance, but in actual applications, in some environmental requirements, the pattern library changes with the changes of external requirements. When the amount of comparison is small When the specified requirements can be met based on the current hardware and software environment, but when the amount of comparison is large and the comparison is based on tens of thousands or more...

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 Patents(China)
IPC IPC(8): G06F17/30G06K9/00
CPCG06F18/22
Inventor 周龙沙邵诗强
Owner TCL CORPORATION
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