Digital Forensics for JPEG2000 and Motion JPEG2000
- When?
- Friday 17 December 2010, 14:00 to 15:00
- Where?
- 39BB02
- Open to:
- Students, Staff
- Speaker:
- Mr Ghulam Qadir
With the advancement of imaging devices and image manipulation soft-ware, production, development and manipulation of digital images can now be done by almost everyone. For this reason, the task of tracking and protecting digital data (images, videos etc.) has become very difficult. To provide adequate policing over the use of digital content, both Active and Passive security techniques are followed. Digital watermarking is an active approach that involves pre-processing an image in order to protect it.
On the other hand, a passive approach, such as Image Forensics utilizes inherent features of an image that are induced by a parent device or a software. This report will focus on one of the tasks of image forensics that is Processing History Recovery specific to JPEG2000(J2K) images. JPEG2000 is a new and improved format introduced by Joint Photographic Experts Group. Unlike JPEG, it is based on DWT and has a complex coding system. Furthermore, the size to compression ratio of JPEG2000 is also better than JPEG. For this reason, it is more suitable for storing large quantity of CCTV data. In this report, we have presented two novel techniques for image forensics. First technique detects the operational history from JPEG2000 images and the second technique extracts images with certain features, such as glare. Since this research revolves around JPEG2000 and its sub formats, such as Motion JPEG2000, Discrete Wavelet Transform will be used as a base
transform in this report. An Empirical law (first digit law) called the Benford's Law is used as a benchmark for analysis as done by Fu et al [14] for JPEG images.
In the first technique, we propose the novel use of Benford's Law for estimating JPEG2000 compression for image forensics applications. The initial results indicate that the 1st digit probability of DWT coefficients follow the Benford's Law. The unknown JPEG2000 compression rates of the image can also be derived and proved with the help of a divergence factor, which shows the deviation between the probabilities and Benford's Law. Based on 1338 test images, the mean divergence for DWT coefficients is approximately 0.0016, which is lower than DCT coefficients at 0.0034. However, the mean divergence for JPEG2000 images compression rate at 0.1 is 0.0108, which is much higher than the uncompressed DWT coefficients. This result clearly indicates a presence of compression in the image. Moreover, we compare the results of the 1st digit probability and divergence among JPEG2000 compression rates at 0.1, 0.3, 0.5 and 0.9. The results acquired through the first technique have been examined in greater detail in the second proposed technique. For 122 images, the Benford's Law graph showed an irregular peak at digit 5, which by definition of the Benford's Law, represents unnatural data. These images are analyzed further to check whether extra features inside an image may have caused this unusual peak since all the images are natural and selected from the UCID [26] database.