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

A Fast and Efficient Approximate Repetitive Image Matching Method

A technology of approximate repetition and matching method, which is applied in the direction of instruments, character and pattern recognition, computer components, etc., can solve the problems of slow extraction speed, large noise, slow image matching speed, etc., achieve noise insensitivity, overcome quantization error, The effect is simple and easy to achieve

Active Publication Date: 2018-05-11
长安通信科技有限责任公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at technical problems such as slow extraction of local feature points, high noise, hard quantization of words, and slow image matching speed in the existing bag-of-words model technology, the purpose of the present invention is to provide a fast and efficient approximate repetitive image matching method

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
  • A Fast and Efficient Approximate Repetitive Image Matching Method
  • A Fast and Efficient Approximate Repetitive Image Matching Method
  • A Fast and Efficient Approximate Repetitive Image Matching Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] In order to make the purpose, technical solutions and advantages of the present invention clearer, the following in conjunction with specific examples, and with reference to the appended figure 1 , to further describe the present invention in detail.

[0035] Execution environment of the present invention adopts a Pentium 4 computer with 2.4G Hz central processing unit and 8G byte internal memory and has compiled the algorithm program of fast and efficient approximate image matching with C++ language, can also adopt other execution environments, not here Let me repeat.

[0036] figure 1 It is a flow chart of a method for approximately repeating image matching in the present invention, comprising the following steps:

[0037] Step 1: Extract the local features of each image in the training image library and perform nonlinear mapping to build a visual vocabulary;

[0038] The local feature mentioned in step 1 refers to the ORB (ORiented Brief) feature, and the ORB feat...

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 fast and efficient approximate repeated image matching method. The method is: 1) extracting the ORB feature of each image in the training image library and carrying out nonlinear mapping to the ORB feature of each image, constructing the visual vocabulary of the training image library; 2) according to the constructed visual vocabulary, Use locally constrained linear coding to sparsely encode the ORB features of each image nonlinear mapping in the training image library; 3) extract the ORB features of the image to be matched and perform nonlinear mapping on it, and then according to the built visual vocabulary Match the ORB features of the non-linear mapping of the image to perform sparse coding; 4) Calculate the sparse coding similarity between the sparse coding of the image to be matched and the sparse coding of the image in the training image library. If the similarity exceeds the set threshold, the matching is successful, otherwise the matching fails. The invention reduces the reconstruction error of the hard quantization method, greatly improves the matching speed, and can be used for real-time matching.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to an approximate image matching (Near-DuplicateImage Match) method. Background technique [0002] With the rapid development of multimedia technology and modern information processing technology, especially the promotion of large-scale mobile applications such as cloud computing, WeChat, and Weibo, the number of images / videos is growing explosively, which inevitably leads to a large number of approximations Repeated images bring great inconvenience to information storage and user retrieval. Therefore, near-duplicate image matching has attracted the interest of some scholars. Through the research of approximate duplicate image matching, on the one hand, image copyright protection can be realized; secondly, in image search engines, it can be used to filter out duplicate images in retrieval results, thereby improving the user’s retrieval quality; in addition, in content-based sensitive...

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): G06K9/64G06K9/46
CPCG06V10/462G06F18/23213
Inventor 李莉戴帅夫刘丙双
Owner 长安通信科技有限责任公司
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