HRDA-Net: image multiple manipulation detection and location algorithm in real scene
Aiming at the problems that the fake image just contains one tampered operation in mainstream manipulation datasets and the artifact is a common problem in manipulation location.The multiple manipulation dataset (MM Dataset) was constructed for real scene, which contained both splicing and removal i...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Journal on Communications
2022-01-01
|
| Series: | Tongxin xuebao |
| Subjects: | |
| Online Access: | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2022016/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841539944350220288 |
|---|---|
| author | Ye ZHU Yilin YU Yingchun GUO |
| author_facet | Ye ZHU Yilin YU Yingchun GUO |
| author_sort | Ye ZHU |
| collection | DOAJ |
| description | Aiming at the problems that the fake image just contains one tampered operation in mainstream manipulation datasets and the artifact is a common problem in manipulation location.The multiple manipulation dataset (MM Dataset) was constructed for real scene, which contained both splicing and removal in each images.Based on this, an end-to-end high-resolution representation dilation attention network (HRDA-Net) was proposed for multiple manipulation detection and localization, which fused the RGB and SRM features through the top-down dilation convolutional attention (TDDCA).Finally, the mixed dilated convolution (MDC) would respectively extract the features of splicing and removal, which could realize multiple manipulation location and confidence prediction.The cosine similarity loss was proposed as auxiliary loss to improve the efficiency of network.Experimental results on MM Dataset indicate that the performance and robustness of HRDA-Net is better than semantic segmentation methods.Furthermore, the scores of F1 and AUC are greater than state-of-the-art manipulation location methods in CASIA and NIST datasets. |
| format | Article |
| id | doaj-art-9164ab04f1a64ac09e4a91963f82e494 |
| institution | Kabale University |
| issn | 1000-436X |
| language | zho |
| publishDate | 2022-01-01 |
| publisher | Editorial Department of Journal on Communications |
| record_format | Article |
| series | Tongxin xuebao |
| spelling | doaj-art-9164ab04f1a64ac09e4a91963f82e4942025-01-14T06:30:33ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2022-01-014321722659398764HRDA-Net: image multiple manipulation detection and location algorithm in real sceneYe ZHUYilin YUYingchun GUOAiming at the problems that the fake image just contains one tampered operation in mainstream manipulation datasets and the artifact is a common problem in manipulation location.The multiple manipulation dataset (MM Dataset) was constructed for real scene, which contained both splicing and removal in each images.Based on this, an end-to-end high-resolution representation dilation attention network (HRDA-Net) was proposed for multiple manipulation detection and localization, which fused the RGB and SRM features through the top-down dilation convolutional attention (TDDCA).Finally, the mixed dilated convolution (MDC) would respectively extract the features of splicing and removal, which could realize multiple manipulation location and confidence prediction.The cosine similarity loss was proposed as auxiliary loss to improve the efficiency of network.Experimental results on MM Dataset indicate that the performance and robustness of HRDA-Net is better than semantic segmentation methods.Furthermore, the scores of F1 and AUC are greater than state-of-the-art manipulation location methods in CASIA and NIST datasets.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2022016/deep learningmultiple manipulation detection and locationMM Datasetcosine similarity loss function |
| spellingShingle | Ye ZHU Yilin YU Yingchun GUO HRDA-Net: image multiple manipulation detection and location algorithm in real scene Tongxin xuebao deep learning multiple manipulation detection and location MM Dataset cosine similarity loss function |
| title | HRDA-Net: image multiple manipulation detection and location algorithm in real scene |
| title_full | HRDA-Net: image multiple manipulation detection and location algorithm in real scene |
| title_fullStr | HRDA-Net: image multiple manipulation detection and location algorithm in real scene |
| title_full_unstemmed | HRDA-Net: image multiple manipulation detection and location algorithm in real scene |
| title_short | HRDA-Net: image multiple manipulation detection and location algorithm in real scene |
| title_sort | hrda net image multiple manipulation detection and location algorithm in real scene |
| topic | deep learning multiple manipulation detection and location MM Dataset cosine similarity loss function |
| url | http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2022016/ |
| work_keys_str_mv | AT yezhu hrdanetimagemultiplemanipulationdetectionandlocationalgorithminrealscene AT yilinyu hrdanetimagemultiplemanipulationdetectionandlocationalgorithminrealscene AT yingchunguo hrdanetimagemultiplemanipulationdetectionandlocationalgorithminrealscene |