内容自适应导引图像滤波器在快速源摄像机识别中的应用

144 浏览量更新于2024-08-29 收藏 453KB PDF 举报

"这篇研究论文探讨了如何使用内容自适应导引图像滤波器(CAGIF)来快速识别源摄像机，即SCI(源相机识别)技术。该技术在图像取证领域具有重要意义，因为能够将图像与拍摄它的特定相机关联起来的关键特征是传感器图案噪声，这是一种估计图像内容与其去噪版本之间的差异。通常认为，基于传感器的SCI性能很大程度上取决于所使用的去噪滤波器。" 文章提出了一种新颖的基于传感器的SCI方法，该方法利用CAGIF的低复杂性特性，大大提高了处理速度，这对于可能的实时SCI应用是一个显著优势。尽管速度提升，但实验结果表明，该方法在准确性方面与当前最先进的方法相当或更优。关键词包括：法医学，图像法医学，源相机识别，传感器图案噪声，去噪。深入解释： 1. 图像法医学：这是一个研究领域，主要关注于通过分析图像来确定其真实性、来源和篡改情况，用于法律调查。 2. 源相机识别（SCI）：这是图像法医学的一个关键部分，其目标是通过分析图像中的特定特征（如传感器噪声）来追踪图像的来源，即确定图像是由哪个相机拍摄的。 3. 传感器图案噪声：这是每个相机传感器特有的随机噪声，源自制造过程中的微小差异。它在图像的高光和阴影区域尤为明显，可以作为识别相机的独特“指纹”。 4. 内容自适应导引图像滤波器（CAGIF）：这是一种图像处理技术，可以根据图像内容自适应地进行去噪，同时尽可能保留图像细节。相比于其他滤波器，CAGIF在保持图像质量的同时，计算复杂度更低，因此更适合快速处理。 5. 去噪：图像处理中消除图像噪声的过程，目的是提高图像清晰度，使分析更准确。在SCI中，有效的去噪是提取传感器图案噪声的关键步骤。 6. 实验结果：论文中的实验展示了新方法在速度和准确性上的优势，证明了CAGIF在SCI中的潜力。这篇研究论文为快速且准确的源相机识别提供了新的思路，利用CAGIF优化了基于传感器的图像指纹提取过程，对图像法医学领域的发展有着积极的推动作用。

TECHNICAL NOTE

DIGITAL & MULTIMEDIA SCIENCES

Hui Zeng,

Ph.D.; and Xiangui Kang,

Ph.D.

Fast Source Camera Identification Using

Content Adaptive Guided Image Filter*

ABSTRACT: Source camera identification (SCI) is an important topic in image forensics. One of the most effective fingerprints for linking

an image to its source camera is the sensor pattern noise, which is estimated as the difference between the content and its denoised version. It

is widely believed that the performance of the sensor-based SCI heavily relies on the denoising filter used. This study proposes a novel sensor-

based SCI method using content adaptive guided image filter (CAGIF). Thanks to the low complexity nature of the CAGIF, the proposed

method is much faster than the state-of-the-art methods, which is a big advantage considering the potential real-time application of SCI. Despite

the advantage of speed, experimental results also show that the proposed method can achieve comparable or better performance than the state-

of-the-art methods in terms of accuracy.

KEYWORDS: forensic science, image forensics, source camera identification, sensor pattern noise, denoising filter, content adaptive guided

image filter

The origin information of a digital image or video is important in

forensic research area as well as in applications; for example in law

enforcement, the investigator would be interested in linking images

of illegal content to a certain camera. One of the most effective fin-

gerprints for linking an image to its source camera is the sensor pat-

tern noise (SPN) (1), which is caused by photo-response

nonuniformity (PRNU) due to inhomogeneity of silicon wafers (2).

An important advantage of SPN is that it can identify not only cam-

era models, but also individual cameras of the same model (3).

In the most widely accepted sensor-based source camera iden-

tification (SCI) method (3), SPN is estimated as the difference

between the content and its wavelet denoised version (4), and its

correlation to the fingerprint of a given camera is used to mea-

sure the probability that an image originated from that camera.

Owing to its wide applications in source identification and integ-

rity verification, this method has been fully studied and various

enhancements have emerged recently. Existing enhancements to

SCI are classifiable in three categories:

1 Better filters are adopted in extracting SPN from the image

(5,6).

2 Better strategies or metrics are adopted to measure the corre-

lation between the SPN from an image and the fingerprint of

a reference camera (7–10).

3 Postprocessing the extracted SPN to further suppress the

unwanted artifacts, for example color interpolation, row and

column artifacts, and contamination of the image details

(7,11).

Our research lies in the first category enhancement of SCI.

Unlike other state-of-the-art SCI methods, in this method, SPN

is extracted from an image using a content adaptive guided

image filter (CAGIF) (12). Thanks to the low complexity nature

of CAGIF, the proposed method is much faster than other the

state-of-the-art methods, which is a big advantage considering

the potential real-time application of SCI. Despite the advantage

of speed, experimental results also show that the proposed

method achieves comparable or better performance than the

state-of-the-art SCI methods in terms of accuracy. In this study,

boldface symbols represent either vectors or matrices.

The rest of the paper is organized as follows. In the next section,

we briefly review of the process of sensor-based SCI and the

denoising filters used in the state-of-the-art literatures. The third

section introduces CAGIF and the proposed CAGIF-based SCI

method. The fourth section shows the experimental results. The

proposed method is compared with other methods in terms of both

accuracy and speed. The conclusion is given in last section.

Brief Review of Source Camera Identification

In this section, we briefly explain the SCI method developed in

ref. (3) and some improved algorithms that relate most to our work.

The PRNU of the imaging sensor is usually modeled as a

multiplicative factor that is unique to a camera A. Taking the

PRNU factor K

into consideration, the raw camera output I can

be written as:

I ¼ I

ð1 þ K

Þþh ð1Þ

where I

is the noise-free image, and h refers to other noise ele-

ments which usually believed to be zero-mean.

School of Information Science & Tech., Sun Yat-sen University, Guangz-

hou 510006, China.

*Supported by NSFC (Grant Nos 61379155 and U1135001), 973 Program

(Grant No. 2011CB302204), the Research Fund for the Doctoral Program of

Higher Education of China (Grant No. 20110171110042), and NSF of

Guangdong Province (Grant No. 2013020012788).

Received 28 Jan. 2015; and in revised form 9 June 2015; accepted 27

June 2015.

JForensicSci, March 2016, Vol. 61, No. 2

doi: 10.1111/1556-4029.13017

Available online at: onlinelibrary.wiley.com

下载后可阅读完整内容，剩余6页未读，立即下载

weixin_38658982

粉丝: 7
资源: 941

内容自适应导引图像滤波器在快速源摄像机识别中的应用

自适应导引图像滤波器，用于改进视频编码中的环路滤波

基于自适应导引图像滤波器和天空区域色彩通道信息的单图像雾霾去除

自适应滤波器 自适应滤波器

用于消除周期性噪声的自适应高斯陷波滤波器：Matlab实现“从数字图像中消除周期性噪声的自适应高斯陷波滤波器”的实现-matlab开发

自适应导引滤波器与高效匹配成本的局部立体匹配算法

自适应形态学滤波器研究

ADSP.rar_ADSP_自适应_自适应滤波器

自适应模拟带通滤波器.docx

efficiznq.rar_Linux/Unix编程_Volterra_稀疏 自适应_稀疏滤波器_自适应滤波器

matlab开发-自适应degree多项式滤波器savitzkygolayfilter

最新资源

自适应滤波器自适应滤波器

efficiznq.rar_Linux/Unix编程_Volterra_稀疏自适应_稀疏滤波器_自适应滤波器