US 12,190,533 B2 · Focal stack camera as secure imaging device and image manipulation detection method

Abstract

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.

Claims7

Dependent claims

An image manipulation detection method, comprising:

receiving, by an image processor, a stack of images for a given scene, where each image in the stack of images is captured concurrently at a different focal plane by an imaging device and one of the images in the stack of images is designated as an image of interest;
determining, by the image processor, whether images in the stack of images are self-consistent across the stack of images;
tagging, by the image processor, the image of interest as authentic in response to the images being self-consistent across the stack of images;
tagging, by the image processor, the image of interest as manipulated in response to the images not being self-consistent across the stack of images;
and
storing, by the image processor, the image of interest along with its tag in a data store.

3 more elements

Dependencies are inferred from the claim text.

Focal stack camera as secure imaging device and image manipulation detection method

Life of this patent

Focal stack camera as secure imaging device and image manipulation detection method

Life of this patent

Abstract

Claims7