基于高斯增强模块的相机模型辨别
Gaussian Enhancement Module for Reinforcing High-frequency Details in Camera Model Identification
黄远航 1边山 2王春桃1
作者信息
- 1. 华南农业大学数学与信息学院 广州 510642;农业农村部华南热带智慧农业技术重点实验室 广州 510642
- 2. 华南农业大学数学与信息学院 广州 510642;农业农村部华南热带智慧农业技术重点实验室 广州 510642;广东省智能信息处理重点实验室深圳市媒体信息内容安全重点实验室 广东深圳 518060
- 折叠
摘要
在多媒体取证中,高通滤波器是卷积神经网络常用的预处理层之一,用于抑制图像内容的影响,只强调高频特征.但与此同时,其他一些包含取证痕迹的有用信息也将被不加区别地剔除.为了解决这一问题,文中提出了一个简单而高效的高斯增强模块(Gaussian Enhancement Module,GEM)来提取"扩展的"高频特征,即在维持原有特征强度的基础上增强高频细节信息.GEM由两个连续的一维低通高斯滤波器组成,以获得一个模糊版本的特征图,并进一步得到相应的扩展高频残差.通过以高频残差作为空间掩膜,它可以自适应地强化脆弱和细微的低级取证特征,并防止在特征传递过程中出现衰减现象.在相机模型辨别数据集上进行实验,通过将该模块插入多个主流骨干网络,GEM仅仅带来非常轻微的模型复杂度的增加,网络性能和鲁棒性却显著提高,表明该模块支持"即插即用",与特定的网络架构无关.
Abstract
In multimedia forensics,a high-pass filter is one of the commonly used pre-processing layers by convolutional neural network to depress the impact of image content and only highlight high-frequency features.However,some other useful informa-tion containing forgery traces would also be removed indiscriminately in the meantime.To address this issue,in this paper,a sim-ple yet effective Gaussian enhancement module is proposed to extract"extended"high-frequency features,namely,reinforce high-frequency details while maintaining the original feature strength.The GEM comprises two successive low-pass Gaussian filters to acquire a blurry version of the feature map and further get the corresponding extended high-frequency residual.It can strengthen fragile and subtle low-level forgery features adaptively and prevent feature attenuation as well.Experiments are conducted on the camera-model identification dataset by plugging the module into several mainstream backbone networks,indicating that it sup-ports"plug and play"and is non-related to the specific network architecture.The proposed GEM brings a significant improve-ment both in the performance and the robustness of networks with the slightly increased complexity of models.
关键词
相机模型辨别/深度学习/图像取证/高通滤波器/高斯增强Key words
Camera model identification/Deep learning/Image forensics/High-pass filter/Gauss enhancement引用本文复制引用
基金项目
广东省智能信息处理重点实验室项目(2023B1212060076)
国家自然科学基金(62172165)
广东省自然科学基金(2022A1515010325)
广州市基础和应用基础研究项目(202201010742)
出版年
2024
NETL
NSTL
万方数据