Developing an Image Manipulation Detection Algorithm Based on Edge Detection and Faster R-CNN
Due to the wide availability of the tools used to produce manipulated images, a large number of digital images have been tampered with in various media, such as newspapers and social networks, which makes the detection of tampered images particularly important. Therefore, an image manipulation detec...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-10-01
|
| Series: | Symmetry |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-8994/11/10/1223 |
| _version_ | 1811187729466130432 |
|---|---|
| author | Xiaoyan Wei Yirong Wu Fangmin Dong Jun Zhang Shuifa Sun |
| author_facet | Xiaoyan Wei Yirong Wu Fangmin Dong Jun Zhang Shuifa Sun |
| author_sort | Xiaoyan Wei |
| collection | DOAJ |
| description | Due to the wide availability of the tools used to produce manipulated images, a large number of digital images have been tampered with in various media, such as newspapers and social networks, which makes the detection of tampered images particularly important. Therefore, an image manipulation detection algorithm leveraged by the Faster Region-based Convolutional Neural Network (Faster R-CNN) model combined with edge detection was proposed in this paper. In our algorithm, first, original tampered images and their detected edges were sent into symmetrical ResNet101 networks to extract tampering features. Then, these features were put into the Region of Interest (RoI) pooling layer. Instead of the RoI max pooling approach, the bilinear interpolation method was adopted to obtain the RoI region. After the RoI features of original input images and edge feature images were sent into bilinear pooling layer for feature fusion, tampering classification was performed in fully connection layer. Finally, Region Proposal Network (RPN) was used to locate forgery regions. Experimental results on three different image manipulation datasets show that our proposed algorithm can detect tampered images more effectively than other existing image manipulation detection algorithms. |
| first_indexed | 2024-04-11T14:08:24Z |
| format | Article |
| id | doaj.art-a9917e408c5b41fb86b3f5512009a5ec |
| institution | Directory Open Access Journal |
| issn | 2073-8994 |
| language | English |
| last_indexed | 2024-04-11T14:08:24Z |
| publishDate | 2019-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Symmetry |
| spelling | doaj.art-a9917e408c5b41fb86b3f5512009a5ec2022-12-22T04:19:49ZengMDPI AGSymmetry2073-89942019-10-011110122310.3390/sym11101223sym11101223Developing an Image Manipulation Detection Algorithm Based on Edge Detection and Faster R-CNNXiaoyan Wei0Yirong Wu1Fangmin Dong2Jun Zhang3Shuifa Sun4College of Computer and Information Technology, China Three Gorges University, Yichang 443002, ChinaCollege of Computer and Information Technology, China Three Gorges University, Yichang 443002, ChinaCollege of Computer and Information Technology, China Three Gorges University, Yichang 443002, ChinaDepartment of Electrical Engineering and Computer Science, University of Wisconsin-Milwaukee, Milwaukee, WI 53201, USACollege of Computer and Information Technology, China Three Gorges University, Yichang 443002, ChinaDue to the wide availability of the tools used to produce manipulated images, a large number of digital images have been tampered with in various media, such as newspapers and social networks, which makes the detection of tampered images particularly important. Therefore, an image manipulation detection algorithm leveraged by the Faster Region-based Convolutional Neural Network (Faster R-CNN) model combined with edge detection was proposed in this paper. In our algorithm, first, original tampered images and their detected edges were sent into symmetrical ResNet101 networks to extract tampering features. Then, these features were put into the Region of Interest (RoI) pooling layer. Instead of the RoI max pooling approach, the bilinear interpolation method was adopted to obtain the RoI region. After the RoI features of original input images and edge feature images were sent into bilinear pooling layer for feature fusion, tampering classification was performed in fully connection layer. Finally, Region Proposal Network (RPN) was used to locate forgery regions. Experimental results on three different image manipulation datasets show that our proposed algorithm can detect tampered images more effectively than other existing image manipulation detection algorithms.https://www.mdpi.com/2073-8994/11/10/1223image manipulation detectionfaster r-cnnedge detectionmax pooling |
| spellingShingle | Xiaoyan Wei Yirong Wu Fangmin Dong Jun Zhang Shuifa Sun Developing an Image Manipulation Detection Algorithm Based on Edge Detection and Faster R-CNN Symmetry image manipulation detection faster r-cnn edge detection max pooling |
| title | Developing an Image Manipulation Detection Algorithm Based on Edge Detection and Faster R-CNN |
| title_full | Developing an Image Manipulation Detection Algorithm Based on Edge Detection and Faster R-CNN |
| title_fullStr | Developing an Image Manipulation Detection Algorithm Based on Edge Detection and Faster R-CNN |
| title_full_unstemmed | Developing an Image Manipulation Detection Algorithm Based on Edge Detection and Faster R-CNN |
| title_short | Developing an Image Manipulation Detection Algorithm Based on Edge Detection and Faster R-CNN |
| title_sort | developing an image manipulation detection algorithm based on edge detection and faster r cnn |
| topic | image manipulation detection faster r-cnn edge detection max pooling |
| url | https://www.mdpi.com/2073-8994/11/10/1223 |
| work_keys_str_mv | AT xiaoyanwei developinganimagemanipulationdetectionalgorithmbasedonedgedetectionandfasterrcnn AT yirongwu developinganimagemanipulationdetectionalgorithmbasedonedgedetectionandfasterrcnn AT fangmindong developinganimagemanipulationdetectionalgorithmbasedonedgedetectionandfasterrcnn AT junzhang developinganimagemanipulationdetectionalgorithmbasedonedgedetectionandfasterrcnn AT shuifasun developinganimagemanipulationdetectionalgorithmbasedonedgedetectionandfasterrcnn |