Image fusion method based on Random Walk

An image fusion, image technology, applied in image enhancement, image data processing, instruments and other directions, can solve the problems of change, color distortion, bad visual effects, etc., to achieve strong ability, strong border color mutation, real image fusion effect. Effect

Active Publication Date: 2013-03-13
ZHEJIANG UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the gradient domain method will cause the color inside the entire patch to change towards the color of the target image on the patch boundary while eliminating the inconsistency on the patch boundary.
Therefore, when the difference between the source image and the target image on the patch boundary is too large, the Poisson image editing method will cause the color distortion of the "salient area" inside the patch, resulting in bad visual effects

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 fusion method based on Random Walk
  • Image fusion method based on Random Walk
  • Image fusion method based on Random Walk

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] First define the abbreviations that will be used in the following instructions:

[0027] : the color depth of the pixels on the target image;

[0028] : the color depth of the pixels on the source image;

[0029] : The patch area intercepted from the source image;

[0030] : The patch is copied to the area on the target map;

[0031] : The position of the outline of the patch on the source image;

[0032] : Copy to the location of the outline of the patch on the target map.

[0033] figure 1 It is the basic flow chart of the present invention. The invention uses simple user interaction in the source image as an input to obtain an output image of a harmonious fusion of the source image patch and the target image. The following describes each process of the present invention in detail:

[0034] 1. Simple user interaction

[0035] In this step, an image patch containing the image to be copied is outlined in the source image, and a specific location on th...

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 provides an image fusion method based on Random Walk. The image fusion method comprises the following steps of: (1) drawing an image patch comprising an image to be copied in a source image, and specifying a specific position where the image patch is to be copied on a target image; (2) constructing an image structure for the image comprising the image to be copied; (3) solving a Dirichlet problem with a Random Walk method to obtain a middle result needed for solving an image fusion problem; and (4) overlaying the middle result generated in the step (3) with the source image, and copying an overlaying result at the copying position on the target image specified in the step (1) to generate an image fusion result. In the method, a user is only required to provide the image patch comprising a 'remarkable object' and specify the copying position on the target image. By adopting the method, boundary inconsistence of the patch is eliminated automatically, and the color of the 'remarkable object' is prevented from changing.

Description

technical field [0001] The invention relates to an image fusion method based on Random Walk. Background technique [0002] Image fusion is an important link in image editing. Its purpose is to copy the objects in the source image to the target image and make the newly generated image visually natural and real. The alpha blend image fusion function provided by commercial image editing tools such as PhotoShop requires the user to completely segment the "salient object" in the source image, and then paste it into the target image. The image segmentation stage requires a lot of user interaction, which is not only time-consuming, but also a small segmentation error will greatly affect the visual effect of image fusion. The Poisson image editing method completely copies the image patch containing "salient objects" in the source image to the target image, and uses the gradient domain method to eliminate the inconsistency between the copied patch and the target image on the patch b...

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): G06T5/50
Inventor 成可立童若峰
Owner ZHEJIANG 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