Image alignment method and its device, computer-readable storage medium and computer equipment

An image alignment and image technology, applied in the field of image processing, to achieve the effect of high alignment accuracy between frames and good image alignment effect

Active Publication Date: 2021-09-17
GUANGZHOU SHIYUAN ELECTRONICS CO LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to solve the above-mentioned technical problem of multi-frame image alignment in a non-planar scene, an embodiment of the present invention proposes an image alignment method and its device, a computer-readable storage medium, and a computer device

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
  • Image alignment method and its device, computer-readable storage medium and computer equipment
  • Image alignment method and its device, computer-readable storage medium and computer equipment
  • Image alignment method and its device, computer-readable storage medium and computer equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] The embodiment of the present invention proposes an image alignment method, such as Figure 1a shown, including the following steps:

[0046] Step S11, obtaining an optical flow information map between two frames of images;

[0047] Step S12, performing superpixel segmentation on the optical flow information map to obtain multiple superpixel block regions;

[0048] Step S13, calculating the homography matrix of each superpixel block region;

[0049] Step S14, according to the homography matrix of each superpixel block region, calculate the map of each superpixel block region;

[0050] Step S15 , extending the edges of the maps of each superpixel block region, and merging to obtain an aligned image.

[0051] The image alignment method proposed in the embodiment of the present invention performs superpixel segmentation on the optical flow information map between two frames of images, and calculates the homography matrix of each sub-region after segmentation, and performs ...

Embodiment 2

[0060] Embodiment 2 of the present invention takes the alignment of two frames of images as an example to further describe the image alignment method and device proposed in Embodiment 1 of the present invention. Two frames of images are color images, one of which is used as a reference image and the other as a matching image.

[0061] The image alignment method proposed in Embodiment 2 of the present invention is as follows Figure 2a shown, including the following steps:

[0062] Step S21, based on the deep learning method, based on the reference image, calculate the optical flow diagram Dx in the x direction and the optical flow diagram Dy in the y direction between two frames of images, and according to the optical flow diagram Dx in the x direction and the optical flow diagram in the y direction Dy, calculate the optical flow information map, where the value of each point on the optical flow information map is the sum of the square of the value of the point in the x-direc...

Embodiment 3

[0074] Embodiment 3 of the present invention, on the basis of Embodiment 2, further describes the image alignment method and device for multi-frame images exceeding two frames.

[0075] The image alignment method proposed in Embodiment 3 of the present invention is as follows Figure 3a shown, including the following steps:

[0076] Step S30, selecting two frames of images from the multiple frames of images, wherein one frame of images is used as a reference image, and the other frame of images is used as a matching image;

[0077] Step S31, based on the deep learning method, based on the reference image, calculate the optical flow diagram Dx in the x direction and the optical flow diagram Dy in the y direction between two frames of images, and according to the optical flow diagram Dx in the x direction and the optical flow diagram in the y direction Dy, calculate the optical flow information map, where the value of each point on the optical flow information map is the sum of...

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

Embodiments of the present invention relate to an image alignment method and device thereof, a computer-readable storage medium, and computer equipment, wherein the method includes: acquiring an optical flow information map between two frames of images; performing superpixel segmentation on the optical flow information map to obtain Multiple superpixel block regions; calculate the homography matrix of each superpixel block region; calculate the mapping map of each superpixel block region according to the homography matrix of each superpixel block region; extend each superpixel block region The edges of the maps are mapped and fused to get aligned images. The image alignment method and its device, computer-readable storage medium and computer equipment proposed in the embodiments of the present invention can achieve better image alignment effects, and the alignment accuracy between frames is high.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an image alignment method and its device, a computer-readable storage medium and computer equipment. Background technique [0002] In application scenarios where multiple frames of images are captured continuously with a single camera to achieve multi-frame synthesis such as night scene image enhancement, multi-frame HDR, and multi-frame image stitching, it is necessary to calculate motion parameters between frames to align multiple frames of images. For a planar image, the motion parameters between frames can be accurately described by a homography matrix; the planar image means that the scene itself is a plane or the scene object is far away from the camera and can be approximated as a plane. For non-planar scenes, the motion parameters between frames cannot be fully described by a homography matrix. For non-planar scenes, there are currently two solutions...

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): G06T5/50G06T7/11
CPCG06T5/50G06T2207/20221G06T7/11
Inventor 邹超洋
Owner GUANGZHOU SHIYUAN ELECTRONICS CO LTD
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