Document images forgery localization using a two‐stream network

Abstract

Document images often contain essential and sensitive information. With image editing software, one can easily manipulate the semantic meaning of the document image by copy-move, splicing, and removal, which causes many security issues. Hence, the document image forgery detection and localization are of great significance. In this paper, a novel two-stream network is proposed to detect and locate the forgery regions of document images. One stream captures forgery traces from the spatial information, including unnatural boundaries, contrast/brightness inconsistencies, and blurring. The other stream eliminates the image semantics by six residual filters, then capture the anomalous features between neighboring pixels introduced by upsampling, downsamping, and inpainting. Finally, the discriminant network is introduced to deeply fuse features from the two streams, and discriminates whether each image patch is forged or not using global average pooling and fully connected layer. Extensive experimental results on our constructed document image data set and the Security AI Challenger Program data set demonstrate the proposed two-stream network outperforms state-of-the-art methods. This proposed two-stream network is also better than each single stream, and is robust to some common forgery postprocessing operations.

Document image forgery detection based on multiscale edge similarity

“…With the increasing application of deep learning techniques in image processing tasks, some researchers [5] have attempted to perform document image forgery detection by constructing deep network models. Xu et al [6] proposed a new dual-stream network architecture that combines residual filters. In this network, the dual streams are responsible for capturing forged traces in the image and abnormal features between adjacent pixels.…”

CRFL: A novel federated learning scheme of client reputation assessment via local model inversion

“…FL cooperates the learning task under the orchestration of a central server while raw client data are shared neither with the server nor among distinct clients 3–5 . With the increasing attention paid to artificial intelligence security and privacy in ML, FL has become a hot research topic over the past half decade 6–13 …”

“…[3][4][5] With the increasing attention paid to artificial intelligence security and privacy in ML, FL has become a hot research topic over the past half decade. [6][7][8][9][10][11][12][13] Unlike traditional centralized ML paradigms, in each round of FL training, the clients train local models on their own data sets, and upload the parameters of the trained model to the server, rather than sharing the raw local data. Then the server aggregates the local models to update the global model and sends it to the clients for the next round of training.…”