Forensic method for identifying forged documents. For each of a stream of incoming jpeg images, using a processor configured for determining whether jpeg image/s is a replacement forgery by determining whether a first portion of individual image which resides at a known location (known likely to be replaced by forger) within the individual jpeg image has been replaced, including: indicator, face-djpg, for the first portion at known location; computing indicator, aka nonface-djpg, for a second portion of individual image which resides at a comparison location within the jpeg image known as unlikely to be replaced by a forger; and determining whether face-djpg and nonface-djpg fulfill predetermined logical criterion and deciding whether the individual jpeg image is a replacement forgery accordingly.

A visual search system facilitates retrieval of provenance information using a machine learning model to generate content fingerprints that are invariant to benign transformations while being sensitive to manipulations. The machine learning model is trained on a training image dataset that includes original images, benign transformed variants of the original images, and manipulated variants of the original images. A loss function is used to train the machine learning model to minimize distances in an embedding space between benign transformed variants and their corresponding original images and increase distances between the manipulated variants and their corresponding original images.

Image security is becoming an increasingly important issue with the progress of deep learning based image manipulations, such as deep image inpainting and deep fakes. There has been considerable work to date on detecting such image manipulations using better and better algorithms, with little attention being paid to the possible role hardware advances may have for more powerful algorithms. This disclosure proposes to use a focal stack camera as a novel secure imaging device for localizing inpainted regions in manipulated images. Applying convolutional neural network (CNN) methods to focal stack images achieves significantly better detection accuracy compared to single image based detection.

A user can wear a device which emits a visual and/or audible output. The output changes over time. A system is capable of predicting the output. Thus, the system can analyze a video and determine, based on observed output of the device, whether the video has been modified. The output can be particularly difficult for humans to modify, detect, understand, or recreate, further impeding attempts to disguise edits to the video.

Methods and systems for identifying altered content are described herein. The system generates a fingerprint for an unverified content item and locates a plurality of content items that match the fingerprint. The system then compares corresponding frames between the unverified content item and each content item of the plurality of content items. The system identifies, based on the comparing, an altered frame in the unverified content item that does not match a corresponding frame in two or more of the plurality of content items. The system also determines that one or more frames of the unverified content item that follow the altered frame match corresponding frames in the two or more of the plurality of content items. The system then generates for display an indication that the unverified content item contains one or more altered frames.

The present invention relates to a forgery detection system and its method for biomedical experiment images, especially for molecular-biological experiment images, such as western blot (WB) and polymerase chain reaction (PCR) results. The forgery detection system mainly comprises a processing module, an image difference computing module, a thresholding module, and an image mixing module are formed in an image analyzing device in the form of a library, a variable or an operand. Moreover, the processing module has a quantization parameter establishing unit, a similar computing unit, and a pseudo background generating unit. The purpose of the image analyzing device is to display an artificial image on the input image.

A method for verifying an image can include: acquiring a first feature point set of a source image and a second feature point set of a target image; determining a target local feature point pair based on the first feature point set and the second feature point set; determining a mapped point of the first feature point on the target image; determining a distance between a second feature point and the mapped point; acquiring a quantity of reference local feature point pairs; and determining that the target image is an image acquired by copying the source image based on the quantity being greater than a target quantity.

Forensic method for identifying forged documents. For each of a stream of incoming jpeg images, using a processor configured for determining whether jpeg image/s is a replacement forgery by determining whether a first portion of individual image which resides at a known location (known likely to be replaced by forger) within the individual jpeg image has been replaced, including: indicator, face-djpg, for the first portion at known location; computing indicator, aka nonface-djpg, for a second portion of individual image which resides at a comparison location within the jpeg image known as unlikely to be replaced by a forger; and determining whether face-djpg and nonface-djpg fulfill predetermined logical criterion and deciding whether the individual jpeg image is a replacement forgery accordingly.

Disclosed is a hidden data removal method based on an autoregressive generative model which is performed by a computer device. The hidden data removal method includes receiving a source image, randomly selecting a target pixel from the source image, and inputting the source image and an identifier of the target pixel to an autoregressive generative model and restoring the target pixel from the source image. The source image is an image in which steganography-based data is hidden, and the autoregressive generative model restores the target pixel on the basis of a pixel value distribution for pixels adjacent to the target pixel in the source image.

Systems and methods herein describe accessing an image, generating a resized image, generating an image feature vector by applying an image classification neural network to the resized image, generating analysis of the image by processing the image feature vector using a machine-learning classifier trained to analyze the image feature vector, and based on the analysis, determining an event that is attributed to the image.