Exposing digital forgeries in color filter array interpolated images

By Alin C. Popescu, Hany Farid. IEEE Transactions on Signal Processing, 2005

702 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/73387b185ea78033ac6ef954e872c02114971f97

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

Citations in the corpus

Determining Image Origin and Integrity Using Sensor Noise (2008, background): “We now explain how the shaping factor and variance are estimated.”
DeepFake Detection by Analyzing Convolutional Traces (2020, background): “Many works try to reconstruct the history of an image[13]; others try to identify the anomalies, such as the study on the analysis of interpolation effects through CFA (Color Filtering Array) [32], analyzing compression parameters [3, 11, 12], etc.”
Noiseprint: A CNN-Based Camera Model Fingerprint (2018, background): “Popular targets of such analyses are lens aberration [5]–[7], camera response function [8]–[10], color filter array (CFA) [11]–[13] or JPEG artifacts [14]–[16].”
What makes fake images detectable? Understanding properties that generalize (2020, background): “A possible solution involves checking for consistency throughout the image – examples include predicting metadata [14] or other low-level artifacts [27,28,29], learning similar embeddings for nearby patches [38], or learning similarity graphs from image patches [25].”
Information Forensics: An Overview of the First Decade (2013, methodology): “In [82], the weighting coefficients from the EM algorithm in [78] and the peak location and magnitudes of the frequency spectrum of the probability map are used as features for classification [82].”
Robust Detection of Region-Duplication Forgery in Digital Image (2006, methodology): “Farid and Popescu developed several statistical methods for detecting forgeries based on region duplication, color filter interpolation, and re-sampling [ 3 - 6 ]. Fridrich presented a method for detecting copy-move type of forgery [7].”
Digital image forgery detection using passive techniques: A survey (2013, methodology): “Natural image model was constructed by Shi et al. (2007a) to detect splicing which consists of statistical features extracted from the test image as well as 2-D arrays generated by applying to the test images multi-size block discrete cosine transform (MBDCT).”
Detecting digital image forgeries using sensor pattern noise (2006, background): “The PRNU noise is not present in completely saturated areas of an image, where all sensor pixels were filled to a full capacity producing a constant signal.”
Source Camera Identification Using Enhanced Sensor Pattern Noise (2009, background): “…and verifying the integrity of images based on the detection of existence or local inconsistencies of device attributes or data processing related characteristics, such as sensor pattern noise (SPN) [1]–[8], camera response function [9], resampling artefacts [10], CFA interpolation artefacts…”
Detecting Digital Image Forgeries by Measuring Inconsistencies of Blocking Artifact (2007, background): “The first attempt is to detect the traces of specific manipulation during forgery creation, such as resampling [1], copypaste [2, 3], and color filter array interpolation [4].”
Digital image forensics via intrinsic fingerprints (2008, methodology): “Once these coefficients are estimated, they are used to reinterpolate the image and find the interpolation error.”
DeepFakes and Beyond: A Survey of Face Manipulation and Fake Detection (2020, methodology): “fingerprints, the analysis of the intrinsic fingerprints introduced by the camera device, both hardware and software, such as the optical lens [27], colour filter array and interpolation [28], [29], and compression [30], [31], among others, and ii) outcamera fingerprints, the analysis of the externa”
Image Forgery Localization via Fine-Grained Analysis of CFA Artifacts (2012, positive): “Regarding the detection of demosaicing traces, Popescu and Farid propose an approach for detecting the interpolation artifacts left on digital images by resampling [15] and de-mosaicing [12] processes.”
Splicebuster: A new blind image splicing detector (2015, background): “Given these pieces of information, one can detect transitions between original and spliced regions, as already suggested back in 2005 [3].”
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],”
Media Forensics and DeepFakes: An Overview (2020, methodology): “distinguish natural images from computer generated images [55], [56], especially after high-pass filtering the image so as to extract more effective features [56], [57].”
Digital image forensics: a booklet for beginners (2010, methodology): “Popescu and Farid [80] use the Expectation/ Maximization algorithm to estimate the probability of each pixel to be correlated with its neighbors: they show how for re-sampled images, these probabilities are arranged in periodic patterns.”