Flat surface detection in photographs for tamper detection
AssigneeSerelay Limited
InventorsRoy AZOULAY, Benjamin HUSTON, Clinton MONTAGUE
Life of this patent
Estimating the term…
Abstract
Photographs can sometimes be altered or changed in order to create photographs which appear to be of a scene which never took place, or did not take place at the place and time when the photograph is purported to have been taken. One way of circumventing known systems to detect this sort of falsified image is to take a photograph of a printed altered image. The present invention is a method of detecting such “photographs of photographs” by comparing two photographs of the same seen taken at the same time, with different exposures.
Claims20
A method of capturing a photograph of a scene and verifying the photograph, the method comprising the steps of:
- using a digital camera to take at least two photographs of the same scene, the photographs being taken at different exposures;
- deriving a luminosity map from each photograph;
- subtracting a luminosity map derived from a first photograph from a luminosity map derived from a second photograph to derive a map of change in luminosity;
- deriving statistical values about the distribution of change in luminosity;
- using the derived statistical values as inputs to at least one machine learning classification model, the machine learning classification model having been trained to identify where the photographs were of flat surfaces;
- and
- outputting an indication as to whether the photograph was of a flat surface.
A system for capturing a photograph, the system including a portable device having a digital camera, and at least one processor, and software adapted to carry out the steps of:
- causing the digital camera to take at least two photographs of the same scene, the photographs being taken at different exposures;
- deriving a luminosity map from each photograph;
- subtracting a luminosity map derived from a first photograph from a luminosity map derived from a second photograph to derive a map of change in luminosity;
- using the derived statistical values as inputs to at least one machine learning classification model, the machine learning classification model having been trained to identify where the photographs were of flat surfaces;
- and
- outputting an indication as to whether the photograph was of a flat surface.
Dependencies are inferred from the claim text.