Local registration-based image splicing method

An image mosaic and partial technology, which is applied in image data processing, graphics and image conversion, instruments, etc., can solve problems such as local registration of areas that cannot be found, and achieve the effect of expanding the range and high robustness

Inactive Publication Date: 2017-05-31
SHANGHAI UNIV
View PDF7 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method can handle some image mosaic problems in the case of large parallax, but when there are a large number of significant image structures in th

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
  • Local registration-based image splicing method
  • Local registration-based image splicing method
  • Local registration-based image splicing method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] The application environment of the embodiment of the present invention is as follows: see figure 2 , the scene of image acquisition is a commercial square, the input two images I 0 and image I 1 , due to the large occlusion at two different viewpoints when shooting, resulting in a large parallax in the picture.

[0048] see figure 1 , an image stitching method based on local registration, including the following steps:

[0049] Step 1: Feature point detection and matching: first perform feature point detection and extract the input image I 0 and I 1 feature points, and match the feature points to obtain N groups of matching points;

[0050] Step 2: Set from the input image I 0 and I 1 The number m of the matching point set on is 0;

[0051] Step 3: Image local registration: the current matching point set m is the local area, and the local area is registered to obtain the corresponding 3×3 homography mapping matrix H m ;

[0052] Step 4: Determine the current ...

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 local registration-based image splicing method. The method comprises the steps of 1) performing feature point detection and matching; 2) setting a number m of a matching feature point pair set from input images I0 and I1 to be 0; 3) performing image local registration; 4) judging whether a registration error of a current local homography mapping matrix Hm is smaller than a registration threshold or is converged or not; 5) adjusting the matching point pair set; and 6) performing image synthesis. The method is an image splicing method aiming at a large-parallax scene and having higher accuracy; the local registration is performed by utilizing a parallax probability; an obtained splicing result is free from structural distortion, ghosting and deformation, and conforms to human eye habits; compared with an existing method, the method is higher in robustness; and the range of scenes processible for an image splicing technology is expanded.

Description

technical field [0001] The invention relates to an image mosaic method, in particular to an image mosaic method based on local registration. Background technique [0002] Image stitching algorithm is the earliest researched and widely used computer vision technology. Image stitching technology is widely used in the fields of video anti-shake, video summarization and panorama generation. For example, the anti-shake function in a video recorder is realized through a registration algorithm between adjacent frames. Today's 360-degree street view and satellite maps on Baidu Maps are also generated through image stitching technology. Image stitching technology mainly includes two parts: image registration and image fusion. Among them, image registration solves the problem of finding overlapping areas between images, which is the core and key of image mosaic technology, while image fusion technology solves the problem of final drawing of registered images, with the purpose of el...

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): G06T3/00G06T3/40
CPCG06T3/0068G06T3/4038
Inventor 李凯汪洋鑫安平沈礼权邹雪妹
Owner SHANGHAI UNIV
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