Image Forensic Tool (IFT): Image Retrieval, Tampering Detection, and Classification

Introduction

In today’s digital world image files play a crucial role. One of the basic forms of information available is in the form of images as it can be easily interpreted by humans. In no time, this kind of information can be shared to various people in different parts of the world. Therefore the integrity of such information is very important. An image file can be easily tampered using various techniques discussed in the later part of this paper. A tampered image can be harmful in various sensitive cases and its range can’t be imagined (Farid, 2006). On the other hand, tampering of any image is much easier than identifying the tampering done. Hence, tamper detection is one of the most required fields to look into in today’s scenario of digital forensics. The only way to detect whether tampering has been performed or not in a digital forensic process is by using image processing techniques (Katzenbeisser, 1999). A proper forensics process is required for verification of such image files (Böhme, 2009). In this paper, the authors have designed a prototype for performing image forensics and implemented the same for the usage of law enforcement. Image forensics is unavoidable in many cases. Modification or tampering of images that are further used for ill-intentions are increasing day by day. Recent such examples (Redi, 2011, p.2) can be seen in Figures 1 and 2.

Figure 1.

The tampered image of Jeffrey Wong Su En while receiving the award from Queen Elizabeth II (W Taktak, 2011)

Figure 2.

The original image of Ross Brawn receiving the order of the British Empire from the Queen (W Taktak, 2011)

Another important example with a different view of image tampering can be seen in Figures 3 and 4.

Figure 3.

Original image of a crime scene (B Soni, 2018)

Figure 4.

Tampered image in which evidence is altered (B Soni, 2018)