Method and apparatus for a natural image model based approach to image/splicing/tampering detection
US8023747B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Feb 9, 2007 |
| Grant date | Sep 20, 2011 |
| Priority date | — |
| Expiry date | Jul 22, 2030 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2201/0083
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Embodiments of the invention are directed toward methods for an effective blind, passive, splicing/tampering detection. The methods of the various embodiments of the invention use a natural image model to detect image splicing/tampering with a model that is based on statistical features extracted from a given test image and multiple 2-D arrays generated by applying the block discrete cosine transform (BDCT) with several different block-sizes to the test images. Experimental results have demonstrated that the new splicing detection scheme outperforms state-of-the-art methods by a significant margin when applied to the Columbia Image Splicing Detection Evaluation Dataset.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.