A multi-branch convolutional neural network for detecting double JPEG compression

TL;DR

This paper introduces a multi-branch CNN that directly uses raw DCT coefficients to detect double JPEG compression, offering an end-to-end forensic tool with improved effectiveness over prior methods.

Contribution

It presents a novel multi-branch CNN architecture that processes raw DCT coefficients for double JPEG detection, enabling end-to-end analysis without pre-processed features.

Findings

Effective detection of double JPEG compression demonstrated

Outperforms previous CNN-based methods

End-to-end detection capability achieved

Abstract

Detection of double JPEG compression is important to forensics analysis. A few methods were proposed based on convolutional neural networks (CNNs). These methods only accept inputs from pre-processed data, such as histogram features and/or decompressed images. In this paper, we present a CNN solution by using raw DCT (discrete cosine transformation) coefficients from JPEG images as input. Considering the DCT sub-band nature in JPEG, a multiple-branch CNN structure has been designed to reveal whether a JPEG format image has been doubly compressed. Comparing with previous methods, the proposed method provides end-to-end detection capability. Extensive experiments have been carried out to demonstrate the effectiveness of the proposed network.