Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Image tampering recognition model training method and device and image tampering recognition method and device

A technology for identifying models and training methods, applied in the field of image recognition, can solve the problems of fraud in the eKYC process, inability to effectively identify tampered images, and low efficiency, so as to reduce the risk of fraud, avoid manual participation, and improve the effect of training.

Active Publication Date: 2020-07-03
ZHONGAN INFORMATION TECH SERVICES CO LTD
View PDF5 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, criminals use software to edit and forge images or alter certificate information, and then upload photos. Due to the limitations of human vision, manual screening methods cannot effectively identify these tampered images, resulting in fraud risks in the eKYC process; in addition, The method of manually screening uploaded images cannot quickly complete the image review, and the efficiency is low

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 tampering recognition model training method and device and image tampering recognition method and device
  • Image tampering recognition model training method and device and image tampering recognition method and device
  • Image tampering recognition model training method and device and image tampering recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0077] Embodiments of the present invention provide a method for training an image tampering recognition model, which can be applied to a server, and the server can be an independent server or a server cluster, such as figure 1 As shown, the method may include:

[0078] Step 101, acquire a training image set and a label of each training image in the training image set, wherein, when the training image is a tampered image, the label of the training image includes tampering position information and tampering type.

[0079] Among them, the training image set is used to train and generate an image tampering recognition model. The training image can be a digital image in various image formats, such as PNG, JPEG, BMP, TIFF format, etc., or it can be extracted from a digital video in various video formats. Key image frames, such as MP4, AVI, MOV, FLV, MKV formats, etc. In this embodiment, the training image set includes real images and tampered images, wherein the tampered images ca...

Embodiment 2

[0138] Based on the image tampering recognition model generated by training in the first embodiment above, the embodiment of the present invention also provides an image tampering recognition method, which can be applied to a server, and the server can be an independent server or server cluster, such as Figure 6 As shown, the method may include:

[0139] Step 601, acquire an image to be recognized.

[0140] In this embodiment, the acquired image to be recognized is scaled according to a fixed size, so as to meet the input image size requirement of the model.

[0141] Step 602, input the image to be recognized into the trained image tampering recognition model for tampering recognition, and obtain the tampering recognition result of the image to be recognized, the tampering recognition result includes the corresponding tampering position information and tampering position information when the image to be recognized is a tampered image Types of.

[0142] Wherein, the trained ...

Embodiment 3

[0154] Based on the first embodiment above, the embodiment of the present invention also provides an image tampering recognition model training device, such as Figure 7 As shown, the device includes:

[0155] Obtaining module 71, is used for obtaining the training image set and the label of each training image in the training image set, wherein, when the training image is a tampered image, the label of the training image includes tampering position information and tampering type;

[0156] Extraction module 72, for taking training image as input, the RGB feature, texture feature and steganographic feature of training image are extracted by the feature extraction network layer of image tampering recognition model to be trained;

[0157] Identification module 73, for inputting the RGB feature, texture feature and steganographic feature of the training image into the detection network layer of the image tampering recognition model, to obtain the tampering recognition result of 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 discloses an image tampering recognition model training method and device, an image tampering recognition method and an image tampering recognition device, and belongs to the technical field of image recognition, and the method comprises the following steps: obtaining a training image set and a label of each training image in the training image set, when the training image is a tampered image, the label of the training image comprising tampered position information and a tampering type; taking the training image as input, and extracting RGB features, texture features and steganography features of the training image through a feature extraction network layer of a to-be-trained image tampering recognition model; inputting the RGB features, the texture features and the steganography features of the training image into a detection network layer of an image tampering recognition model, and obtaining a tampering recognition result of the training image; and performing difference comparison on the tampering recognition result and the label, optimizing parameters of the image tampering recognition model according to a difference comparison result, continuing training, and ending the training until a predetermined condition is met. According to the embodiment of the invention, the tampered image can be accurately and quickly recognized.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to an image tampering recognition model training method, an image tampering recognition method and a device. Background technique [0002] eKYC (electronic Know Your Customer) refers to the completion of review and filing through electronic means, and has a wide range of applications in scenarios such as identity verification, account opening, and anti-fraud. In the eKYC process, the integrity authentication of the uploaded image is required to determine whether the image has been tampered with. [0003] At present, when authenticating images, most of them use manual methods to check whether the uploaded images are falsified images. However, criminals use software to edit and forge images or alter certificate information, and then upload photos. Due to the limitations of human vision, manual screening methods cannot effectively identify these tampered images, resulting in...

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): G06F21/64G06K9/62G06K9/46G06N3/04G06N3/08
CPCG06F21/64G06N3/084G06V10/56G06N3/045G06F18/241G06F18/214
Inventor 谢畅钱浩然王恒袁皓
Owner ZHONGAN INFORMATION TECH SERVICES CO LTD
Features
  • Generate Ideas
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com