Unlock instant, AI-driven research and patent intelligence for your innovation.

Double-flow network image forgery detection method and system based on image block feature extraction

A feature extraction and image detection technology, applied in the field of computer vision, can solve problems such as not considering the impact of classification, and achieve the effect of less training samples, high accuracy, and good face forgery detection

Active Publication Date: 2021-09-07
UNIV OF SCI & TECH OF CHINA
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Existing face forgery detection inventions seldom analyze from the perspective of local image blocks, and even less explore from the spatial neighborhood information of local image blocks, and do not consider the impact of different pixels in local space on classification

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
  • Double-flow network image forgery detection method and system based on image block feature extraction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] In today's picture forgery technology has been able to generate more and more realistic false face images, the purpose of the present invention is to design a set of face forgery detection scheme, which can detect and distinguish real face images and false faces generated by various means image.

[0037] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0038] At present, the existing research on face forgery detection is all analyzed from the complete face image, seldom analyzed from the perspective of local image blocks, and even less is explored from the spatial neighborhood information of local image blocks. None of the existing methods consider the impact of different pixels in local space on forgery detection.

[0039] In view of the above problems, the present in...

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 a double-flow network forged image detection method based on local image block feature extraction. The method comprises the following steps: carrying out image block segmentation on a to-be-detected image; respectively inputting the obtained image blocks into a CNN model and a CNN-GRU model to extract artifacts in the image blocks and spatial features between the image blocks, obtaining corresponding forgery scores respectively, wherein the closer the forgery scores are to 1, the higher the forgery possibility of the to-be-detected image is; and fusing the two obtained forgery scores by adopting an attention-based fusion method to obtain a score for judging the authenticity of the to-be-detected image.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a double-stream network forged image detection method and system based on image block feature extraction. Background technique [0002] In recent years, the rapid development of artificial intelligence technology and the rise of deep learning technology have produced a large number of related technologies in the field of image and video generation. People have been able to use a variety of technologies to generate fake pictures and videos that can deceive the human eye. Although this type of generation technology can bring some interesting or useful applications in some occasions, maliciously using and disseminating these falsely generated pictures to create some public opinion (such as judicial evidence collection, news reports, medical identification, pornographic dissemination), It will bring some troubles and panic to the society, and even bring some political security issues....

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): G06K9/00G06K9/46G06N3/04G06N3/08
CPCG06N3/08G06N3/048G06N3/045
Inventor 章雪琦谢海永吴曼青
Owner UNIV OF SCI & TECH OF CHINA