Method for positioning forged region based on two view geometry and image division

A graph segmentation and region technology, applied in the field of digital image authenticity identification and image processing, can solve the problems of inapplicability, long training time, and large training samples in the field of forgery region detection

Inactive Publication Date: 2010-01-20
TIANJIN UNIV
View PDF0 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But this method requires larger training samples and longer training time
And this method also requires manual intervention by the user, so it is not applicable in the field of fake area detection

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
  • Method for positioning forged region based on two view geometry and image division
  • Method for positioning forged region based on two view geometry and image division
  • Method for positioning forged region based on two view geometry and image division

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] The present invention is based on two-view geometry and graph segmentation precise location technology of forged areas, adopts two-view geometric constraints to obtain a binary map of forged areas, and utilizes online feature / parameter selection technology to accurately locate forged areas. figure 1 It is a system flow chart of the present invention.

[0043] In the following, the method of obtaining the binary map of the forged region by using the two-view geometric constraint will be described in detail first. The specific steps are as follows, see Figure 5 :

[0044] In the first step, for the obtained video frame or image pair, use planar homography constraint: H constraint to judge the authenticity of the image.

[0045] First of all, for the obtained image pair, first extract the corresponding feature points, then calculate the planar homography matrix H, and then restore one image to another image by the calculated H matrix, and finally perform pixel-level dif...

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 authenticity identification of digital image and image processing, relating to a method for precisely positioning a forged region based on two view geometry and image division. The method is on the basis of two-view geometry theory in camera imaging process, respectively uses strong geometric constraint (H constraint) and weak geometric constraint (F constraint) as feature constraints to accurately evaluate the authenticity of the image through online selection and map segmentation techniques. The invention can carry out identification when only a plurality of images with overlapped parts are found, thereby having very wide applicability.

Description

technical field [0001] The invention belongs to the technical field of digital image authenticity identification and image processing, and in particular relates to a method for detecting forgery of a picture. Background technique [0002] The accurate positioning of the counterfeit area has a wide range of applications in both military and civilian applications, and it also has great application potential in the security field. The present invention uses image pairs containing the same scene to estimate the exact location of the faked area in the photo. Its input is: (1) image pairs containing the same scene but with different camera centers, or (2) image pairs with concentric cameras or scenes coplanar, such as figure 1 shown. The output of the inventive technique is: the exact location of the faked area in the photo pair. [0003] The background technologies used in the present invention are: (1) Robert T. Collins and Yanxi Liu et al [1] proposed an online selection tec...

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 Applications(China)
IPC IPC(8): G06T7/00
Inventor 操晓春曲彦龄张炜王秀锦
Owner TIANJIN 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

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