Hybrid features and semantic reinforcement network for image forgery detection

Chen, Haipeng; Chang, Chaoqun; Shi, Zenan; Lyu, Yingda

doi:10.1007/s00530-021-00801-w

Cited by 19 publications

(10 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This model is extremely precise in locating size or shape of copy-moved areas. Marra et al 17 20 devised a hybrid features and semantic reinforcement network (HFSRNet) for image forgery detection. For the purpose of capturing traces from the image patches and identifying manipulation artifacts, long-short term memory with resampling characteristics has been implemented.…”

Section: Literature Surveymentioning

confidence: 99%

Squirrel Henry Gas Solubility Optimization driven Deep Maxout Network with multi‐texture feature descriptors for digital image forgery detection

Nirmala Priya,

Suresh Kumar,

Suganthi

et al. 2024

Concurrency and Computation

View full text Add to dashboard Cite

SummaryThe ability to detect image forgeries is a crucial component in solving many problems, especially social ones that arise in legal proceedings, on Facebook, and so forth. Generally, forgery detection attempts to predict the artifacts through evaluating the differences in texture properties of an image. One division of an image is copied in a comparable image, typically at a different position, in a process known as copy move forgery. The two main types of copy move forgery prediction are keypoint‐based and block‐based techniques. While the block‐based approach takes more time, the keypoint‐based model performs poorly. The goal of this research was to develop a method for detecting image counterfeiting called the Squirrel Henry Gas Solubility Optimization algorithm‐based Deep Maxout Network (SHGSO‐based DMN). Here, Steerable Pyramid Transform (SPT) decomposition is done in order to obtain quantity of multi‐scale and multi‐oriented sub bands. For the purpose of detecting forgeries, the Deep Maxout Network which was constructed using the SHGSO model is deployed. With accuracy of 0.9507, True Negative Rate (TNR) of 0.9568, and True Positive Rate (TPR) of 0.9517, the new method outperformed other current forgery detection techniques.

show abstract

Section: Literature Surveymentioning

confidence: 99%

Squirrel Henry Gas Solubility Optimization driven Deep Maxout Network with multi‐texture feature descriptors for digital image forgery detection

Nirmala Priya,

Suresh Kumar,

Suganthi

et al. 2024

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…In the massive data analysis scenario, the end-to-end forensics model simplifies the complex process of the analysis model and makes data processing more efficient. In recent years, with the success of deep learning in various computer vision tasks, such as object detection [5,10] and semantic segmentation [11,12], many methods based on deep learning have been developed for image tampering detection and localization [13], the efficiency of neural network models in massive data processing and the accuracy of digital image fingerprint extraction have been continuously highlighted. Therefore, a large number of multimedia forensics analyses began to adopt end-to-end forensics framework based on the deep learning models.…”

Section: Related Workmentioning

confidence: 99%

Image Tampering Localization based on Pyramid Multi-Scale Pooling Module

2022

Proceedings of WCSE 2022 Spring Event: 2022 9th International Conference on Industrial Engineering and Applications

View full text Add to dashboard Cite

In this paper, we proposed a new two-stream image forgery localization method based on pyramid multi-scale pooling module. One stream learns the differences in JPEG compression rate between original regions and tampered regions by processed through ELA. In the second stream, the network learns the global image manipulation features from the input image RGB values. Next the fusion stage fuses multiple tampering features from the dual-channel feature extraction network to generate the shallow feature map, Coordinate Attention is applied to refine the predicted mask and further improve the localization accuracy. And then through the multi-scale pyramid pooling module, so the secondary fusion is carried out. Finally, the fused feature map restores the original image size by using UpSampling for Pixel level classification prediction. The proposed net can effectively decrease incorrect prediction since it makes better use of the contextual spatial information in images. And most of all, from two stages extraction fuses lowlevel and high-level information to refine the global manipulation features. The experimental results on several public datasets show that the proposed model outperforms some state-of-the-art methods.

show abstract

“…Bappy, et al [13] presented a hybrid CNN-LSTM model to learn the spatial structure between those tampered and and non-tampered regions in the shared boundary. Chen et al [14] proposed an encoding and decoding framework that employes both hybrid features and semantic reinforcement network (HFSRNet) for image forgery detection. Particularly, resampled features with longshort term memory is utilized to capture traces from the image patches for finding manipulating artifacts.…”

Section: Introductionmentioning

confidence: 99%

“…Particularly, resampled features with longshort term memory is utilized to capture traces from the image patches for finding manipulating artifacts. However, all the above-mentioned methods [12], [13], [14] perform image tamper detection by processing pictures in patch first, which means they focus only on local areas and ignore relationships between sub-regions, thus their localization performance depends on the size of the patches.…”

Section: Introductionmentioning

confidence: 99%

TransU²-Net: A Hybrid Transformer Architecture for Image Splicing Forgery Detection

Yan

Li³

2023

IEEE Access

View full text Add to dashboard Cite

In recent years, various convolutional neural network (CNN) based frameworks have been presented to detect forged regions in images. However, most of the existing models can not obtain satisfactory performance due to tampered areas with various sizes, especially for objects with largescale. In order to obtain an accurate object-level forgery localization result, we propose a novel hybrid transformer architecture, which exhibits both advantages of spatial dependencies and contextual information from different scales, namely, TransU 2 -Net. Specifically, long-range semantic dependencies are captured by the last block of encoder to locate large-scale tampered areas more completely . Meanwhile, nonsemantic features are filtered out by enhancing low-level features under the guidance of high-level semantic information in the skip connections to achieve more refined spatial recovery. Therefore, our hybrid model can locate spliced forgeries with various sizes without requiring large data set pre-training. In comparison with other existing CNN-based methods, our framework achieves better performance over state-of-the-art methods.

show abstract

Hybrid features and semantic reinforcement network for image forgery detection

Cited by 19 publications

References 31 publications

Squirrel Henry Gas Solubility Optimization driven Deep Maxout Network with multi‐texture feature descriptors for digital image forgery detection

Squirrel Henry Gas Solubility Optimization driven Deep Maxout Network with multi‐texture feature descriptors for digital image forgery detection

Image Tampering Localization based on Pyramid Multi-Scale Pooling Module

TransU²-Net: A Hybrid Transformer Architecture for Image Splicing Forgery Detection

Contact Info

Product

Resources

About

Hybrid features and semantic reinforcement network for image forgery detection

Cited by 19 publications

References 31 publications

Squirrel Henry Gas Solubility Optimization driven Deep Maxout Network with multi‐texture feature descriptors for digital image forgery detection

Squirrel Henry Gas Solubility Optimization driven Deep Maxout Network with multi‐texture feature descriptors for digital image forgery detection

Image Tampering Localization based on Pyramid Multi-Scale Pooling Module

TransU2-Net: A Hybrid Transformer Architecture for Image Splicing Forgery Detection

Contact Info

Product

Resources

About

TransU²-Net: A Hybrid Transformer Architecture for Image Splicing Forgery Detection