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

Image Copy-Paste Tampering Detection Method Based on Segmentation and Deep Convolutional Networks

A deep convolution and tamper detection technology, which is applied in image analysis, image data processing, biological neural network models, etc., can solve the problems of poor tamper edge detection and loss of image detail information, and achieve high tamper detection accuracy Effect

Active Publication Date: 2022-03-18
SHANDONG UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the loss of image detail information after depthwise convolution, the accuracy of detection results of existing methods, especially the detection of tampered edges is not good.

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 Copy-Paste Tampering Detection Method Based on Segmentation and Deep Convolutional Networks
  • Image Copy-Paste Tampering Detection Method Based on Segmentation and Deep Convolutional Networks
  • Image Copy-Paste Tampering Detection Method Based on Segmentation and Deep Convolutional Networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Embodiment 1 of the present disclosure provides an image copy-paste tampering detection method based on segmentation and deep convolutional network.

[0045]Using image segmentation technology, the image is irregularly divided into blocks in advance. Due to the characteristics of copy-paste tampering, the copied area is highly similar to the tampered area and will be divided into the same type of block. Therefore, pre-segmentation processing and strengthening the connection within the same type of block can improve the matching accuracy. Using deep convolutional network for image feature extraction, automatic learning features are used to replace traditional manual features, which makes the extracted features more suitable for image copy-paste tampering detection and avoids the limitations of manual feature extraction. During feature extraction, a feature pyramid structure is constructed, taking into account multi-scale information and detailed information, so that the ...

Embodiment 2

[0103] Embodiment 2 of the present disclosure provides an image copy-paste tampering detection system based on segmentation and deep convolutional network, adopting the image copy-paste tampering detection method based on segmentation and deep convolutional network provided in Embodiment 1, include:

[0104] An image acquisition module, configured to acquire an image to be detected;

[0105] Segmentation module, builds image segmentation model, trains described image segmentation model, obtains segmentation weight parameter and the boundary pixel direction information of image, carries out segmentation processing to described image to be detected, obtains segmented image;

[0106] A feature extraction module, based on a deep convolutional network, performs feature extraction of the image to be detected, and outputs image features;

[0107] The autocorrelation matching module combines the segmented image and the image features to perform autocorrelation matching to obtain imag...

Embodiment 3

[0110] Embodiment 3 of the present disclosure provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the image copy-paste based on segmentation and deep convolutional network as described in Embodiment 1 of the present disclosure is implemented. Steps in a tamper detection method.

[0111] The detailed steps are the same as the image copy-paste tampering detection method based on segmentation and deep convolutional network provided in Embodiment 1, and will not be repeated here.

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 present disclosure provides an image copy-paste tampering detection method based on segmentation and deep convolutional network, comprising the following steps: acquiring an image to be detected; building an image segmentation model, performing training and segmentation processing on the image to be detected, and obtaining the segmentation The weight parameter and the boundary pixel direction information of the image are obtained to obtain a segmented image; the feature extraction of the image to be detected is performed based on a deep convolutional network, and the image feature is output; combined with the segmented image and the image feature, autocorrelation matching is performed to obtain Image matching feature; input the image matching feature into the classification model to obtain a preliminary tampered area detection image; extract the edge information image through the boundary pixel direction information of the image; build a detail optimization model, input the preliminary tampered area detection image and the obtained The above edge information image is output to tamper detection image.

Description

technical field [0001] The disclosure belongs to the technical field of image tampering detection, and in particular relates to an image copy-paste tampering detection method based on segmentation and deep convolutional network. Background technique [0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art. [0003] With the popularization and easy operation of image editing software, tampering with images emerges in an endless stream. If tampered images appear in news media, medical diagnosis, military reconnaissance, judicial evidence collection and other fields, it will definitely bring huge hidden dangers to information security and seriously threaten social harmony and political stability. Therefore, the authenticity identification of images is very important. Copy-paste tampering detection is an important direction of authenticity identification, and the image tampering s...

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): G06T7/00G06V10/26G06V10/44G06V10/75G06V10/774G06V10/82G06K9/62G06N3/04
CPCG06T7/0002G06N3/045G06F18/22G06F18/214
Inventor 王成优李倩雯周晓
Owner SHANDONG UNIV