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

Frequency domain shape description method for image matching, recognition and retrieval

A shape description and image technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve problems such as high computational complexity, robustness to noise and deformation, and inability to obtain tangent vectors, achieving a wide range of application prospects, computing low cost effect

Inactive Publication Date: 2014-03-26
FUDAN UNIV
View PDF1 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Shape context is a feature that is more robust to noise interference, but the shape context feature does not have rotation invariance. It is necessary to use the tangent vector of the image gray-scale domain where the feature point is located as a reference direction to obtain rotation invariance, but this The rotation invariance dependent on external conditions is not robust enough, especially for binary images, it is impossible to obtain a tangent vector that can be used as a reference direction
Therefore, the inventor has obtained the authorized Chinese invention patent [Yang Su: A general feature description method for symbol recognition, invention patent, authorized time: February 6, 2008, authorized country: China, authorized number : 200410016733.0] and paper [S. Yang: "Symbol recognition via statistical integration of pixel-level constraint histograms: A new descriptor", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.27, No.2, pp.278-281 ,2005] proposed a point-constrained histogram feature, which has inherent rotation and scaling invariance, and is more robust to noise and deformation, but the disadvantage is that the computational complexity is high, about O(N 3 )

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
  • Frequency domain shape description method for image matching, recognition and retrieval
  • Frequency domain shape description method for image matching, recognition and retrieval
  • Frequency domain shape description method for image matching, recognition and retrieval

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0019] Step 1: Extract feature points from an input image, and calculate the shape descriptor of each feature point, let P={P 1 ,P 2 ,...,P K} and {f(F(P k ))|k=1,2,…,K} represent the obtained feature points and their corresponding shape descriptors respectively. The calculation steps of shape descriptors are as follows:

[0020] (a) Choose a feature point P k ∈P is used as a reference point to make statistics on the spatial distribution of other feature points to obtain a corresponding histogram, denoted as h(P k ); here, the specific calculation method of the histogram is as follows: with reference point P k As the center, divide the space where the smallest circumscribed circle of the image is into a grid of M×N, calculate the number of feature points falling into each interval of the grid to obtain a histogram, and M and N are all natural numbers; As a reference point, a histogram is obtained corresponding to each feature point, and a total of K histograms {h(P k )|k...

Embodiment 2

[0065] Step 1: same as step 1 of embodiment 1;

[0066] Step 2: All the feature points P={P obtained in step 1 1 ,P 2 ,...,P K} corresponding shape descriptor [f(F(P 1 )), f(F(P 2 )),…,f(F(P K ))] matrix addition to get G(P)=f(F(P 1 ))⊕f(F(P 2 )),…,⊕f(F(P K )), where the operator "⊕" represents the addition of the elements of the corresponding positions of each matrix involved in the operation, and G(P) represents all K matrices f(F(P 1 )), f(F(P 2 )),…,f(F(P K )) The elements with the same subscript are added together, and G(P) is the shape descriptor that is finally used for image matching, recognition, and retrieval;

[0067] Step 3: same as step 2 of embodiment 1;

[0068] Step 4: All the feature points Q={Q obtained in step 3 1 ,Q 2 ,...,Q L} corresponding shape descriptor f(F(Q 1 )), f(F(Q 2 )),…,f(F(Q L )) matrix addition to get G(Q)=f(F(Q 1 ))⊕f(F(Q 2 )),…,⊕f(F(Q K ));

[0069] Step 5: Straighten the matrices G(P) and G(Q) into vectors respectively...

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 belongs to the technical field of pattern recognition, image processing and computer vision, and particularly relates to a frequency domain shape description method for image matching, recognition and retrieval. According to the frequency domain shape description method for the image matching, recognition and retrieval, an image region is divided into a plurality of sub regions, each feature point is used as a reference point to count the number of other feature points falling upon each sub region to obtain a column diagram, a power spectrum of the column diagram is calculated to be used as a shape description, the similarity degree between the shape descriptors of two images is calculated to obtain the similarity degree of the images, and therefore the image matching, recognition and retrieval are achieved. Experiments show that the shape context spectrum signatures of the frequency domain shape description method for the image matching, recognition and retrieval has good rotation and stretching invariance properties, and good robustness on noise and deformation, and is low in calculation cost.

Description

technical field [0001] The invention belongs to the technical fields of pattern recognition, image processing and computer vision, and in particular relates to an image shape feature extraction method, which can be used for image matching, recognition and retrieval. Background technique [0002] Images include shape, texture, and color information. Among them, shape is the main information that image recognition and retrieval rely on. Shape feature description is very important for image recognition and retrieval. [S.Belogie, J.Malik, J.Puzicha: "Shape matching and object recognition using shape contexts, IEEE Transactions on Pattern Analysis and Machine Intelligence”, Volume 24, pp.509-52, 2002] The paper proposes a shape description method called Shape Contexts (Shape Contexts), which first extracts the image Feature points (edge ​​points), and then construct a grid with each feature point as a reference point, count the number of feature points contained in each area of ​...

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
IPC IPC(8): G06K9/62
CPCG06V10/462
Inventor 杨夙
Owner FUDAN UNIV
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