SIFT mismatching-point elimination method based on function fitting

A technology of mismatching points and function fitting, applied in the field of image processing, can solve the problems of low matching efficiency and time-consuming mismatching, and achieve the effects of improving matching efficiency, processing speed and matching accuracy.

Inactive Publication Date: 2018-06-29
XI'AN POLYTECHNIC UNIVERSITY
View PDF5 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a SIFT error matching point elimination method based on function fitting, which solves the problems of time-consuming error matching and low matching efficiency in the prior art

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
  • SIFT mismatching-point elimination method based on function fitting
  • SIFT mismatching-point elimination method based on function fitting
  • SIFT mismatching-point elimination method based on function fitting

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0074] There are 37 key points in the present invention, i.e. n=37, that is, according to the calculation result of formula (1), the function model selected by the image to be matched is a cubic polynomial, and the function model coefficient is obtained according to the least square method, and the minimum square Multiplication is also known as the minimum average method. Its basic idea is to find the best function match of the data by minimizing the square sum of the errors, and to make all the data satisfy the function parameters to the greatest extent possible; the realization is as follows

[0075] Select a set of data r 0 (x), r 1 (x), r 2 (x), r 3 (x), 3

[0076] f(x)=a 0 r 0 (x)+a 1 r 1 (x)+a 2 r 2 (x)+a 3 r 3 (x)

[0077] Among them, a 0 ,a 1 ,a 2 ,a 3 is an undetermined coefficient.

[0078] Make the 37 points to match (x i ,y i ) and the curve y=f(x) have the smallest sum of squares of distance, record

[0079]

[0080] Then find...

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 an SIFT mismatching-point elimination method based on function fitting. Firstly, an SIFT algorithm is adopted to directly match a to-be-matched image and a matching image; thena function model is constructed according to key points of the to-be-matched image after matching, and fitting is carried out with feature key-points of the matching image; and finally, errors of thetwo parts are calculated, and when an error is greater than a given threshold value, it is confirmed that a point thereof is a mismatching point, and the same is eliminated. Experiment results show that compared with an RANSAC algorithm, adopting of function-based fitting to eliminate mismatching points saves 2s on time, increases a correct-matching rate by 15.2%, and retains correct-matching points more. The method solves the problems of mismatching, time consuming proneness and low matching efficiency existing in the prior art.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a function fitting-based SIFT error matching point elimination method. Background technique [0002] Image matching is a key technical issue in image processing. The task of image matching is to find the correspondence between feature points in two or more images of the same scene; the purpose is to obtain similar areas in two images after matching. And determine the point of the same name. Image matching has been extended to many industries, and its development has been greatly improved. It has a wide range of practical applications in many fields, such as image stitching and fusion, target recognition and tracking, photogrammetry remote sensing, image retrieval, etc. In recent years, there have been many image matching algorithms, especially the scale-invariant feature transformation (SIFT) and a variety of SIFT improved algorithms. Cheng Dezhi et al. prop...

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): G06K9/46G06K9/62
CPCG06V10/462G06V10/751
Inventor 李云红钟晓妮黄梦龙袁巧宁张震宇贾凯莉成中豪
Owner XI'AN POLYTECHNIC 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