Determining Image Origin and Integrity Using Sensor Noise

By Mo Chen, Jessica J. Fridrich, Miroslav Goljan, Jan Lukás. 2008

923 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Faceless Person Recognition: Privacy Implications in Social Media (2016, background): “Some works have shown that it is possible to identify the camera taking the photos (and thus link photos and events via the photographer), either from the file itself [7] or from recognisable sensing noise [8, 9].”
Detection of AI-Generated Synthetic Faces (2022, background): “Device ﬁngerprints have been ﬁrst exposed in the seminal work of Chen et al. [34] and Lukas [9].”
Using decision fusion of feature selection in digital forensics for camera source model identification (2012, background): “Because PRNU cannot be easily removed from digital cameras due to costs, [9,26] argued that PRNU is the ideal biometric for digital cameras.”
Autoencoder with recurrent neural networks for video forgery detection (2017, background): “Moreover, the performance is very sensitive to the scene content, limiting application in dark or textured areas [16].”
Information Forensics: An Overview of the First Decade (2013, background): “A NeymanPearson rule for identifying falsified image regions can then be designed to determine the optimal detection threshold for a desired false alarm rate [103].”
An Overview on Image Forensics (2013, methodology): “In [23], controlled cameraspeci�c training data is used to obtain a maximum likelihood PRNU predictor.”
Do GANs Leave Artificial Fingerprints? (2018, background): “Such fingerprints can be used to perform image attribution [11], [13], as well as to detect and localize image manipulations [12], [13], and represent one of the strongest tools in the hands of the forensic analyst.”
CASIA Image Tampering Detection Evaluation Database (2013, methodology): “Similarly, in [11], the method they proposed for forgery detection was also tested on their own collected gray image database which is not available to other researchers.”
Image Splicing Localization using a Multi-task Fully Convolutional Network (MFCN) (2017, background): “The ﬁrst class of splicing localization algorithms exploits noise patterns under the assumption that diﬀerent images have diﬀerent noise patterns as a result of a combination of diﬀerent camera makes/models, the capture parameters of each image, and post-processing techniques [26, 27, 9, 7, 8, 21].”
Domain Fingerprints for No-Reference Image Quality Assessment (2019, positive): “Prior digital ﬁngerprint techniques focus on detecting hand-crafted features for device ﬁngerprints [25].”
Forensic detection of image manipulation using statistical intrinsic fingerprints (2010, methodology): “While each of these methods possesses their own limitations, it has been posited that if a large set of forensic methods are developed, it will be difficult for a forger to create an image capable of fooling all image authentication techniques [1].”
Digital Image Authentication From JPEG Headers (2011, background): “…[5], [6]; color filter array artifacts [7], [8]; inconsistencies in camera response function [9]; inconsistencies in lighting and shadows [10]–[13]; inconsistencies in chromatic aberrations [14], [15]; inconsistencies in sensor noise [16], [17]; and inconsistencies in statistical features [18].”
Splicebuster: A new blind image splicing detector (2015, background): “When either the host camera or an arbitrary number of images taken from it are available, one can estimate the so-called camera fingerprint, or photo-response non-uniformity noise (PRNU) pattern [1].”
Region Duplication Detection Using Image Feature Matching (2010, background): “Other intrinsic cues from the imaging process, including color filter array interpolation patterns [44], [46], camera response functions [30], and camera sensor noise [9], [18], or general statistical properties of untampered natural images [34],”
Noiseprint: A CNN-Based Camera Model Fingerprint (2018, methodology): “PRNU-based methods have shown excellent performance for source identification [29] and for image forgery detection and localization [28], [30]–[33].”
Source Camera Identification Using Enhanced Sensor Pattern Noise (2009, methodology): “While many methods [9]–[12] require that specific assumptions be satisfied, methods based on SPN [1]–[8], [20]–[22] have drawn much attention due to the relaxation of the similar assumptions.”
Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints (2018, background): “Previous work on digital fingerprints is limited to device fingerprints [42, 21] or in-camera post-processing fingerprints [24], which cannot be easily adapted to GAN-generated images.”
Digital image forensics: a booklet for beginners (2010, methodology): “In a later study, Chen and others [16, 18] propose to refine the previous method by estimating PRNU information from pattern noise W.”
Media Forensics and DeepFakes: An Overview (2020, background): “[101] a predictor is designed to adapt the decision threshold to the local image statistics, while in [102] disturbing non-”