Multimedia Forensics

BY Ruchika Diwedi | September 03, 2020

Multimedia Forensics

Probably one among the foremost exciting and relevant fields within the computing world these days is forensics. Like its name, computer forensics is the process  of assembling, analyzing, and reporting on digital evidences so that it can be made both legal and admissible in court.  Utilizing forensics to the multimedia is an evidence-based method which will be used for the revelation as well as   preclusion of a cybercrime or any  such occurrence that involves the mishandling of data. However instead of blood spatters or deoxyribonucleic acid (DNA) swabs, computer forensics depends on proof within the variety of emails, files, online documents, and digital fingerprints. virtual evidences from such incidents of modern age can be used to  disclose the engagements and intentions of hackers in the same way as the physical evidences leads the investigators to the criminal in real world. During this method, computer forensics permits the extension of law enforcement into the digital spheres, additionally referred to as the place where people spend their major time . such  contemporary shift has allowed law enforcement personals  to deal with web crimes as robustly as real-world crimes and to use digital evidence as a legitimate means that for implementing cyber laws and prosecuting people who abuse them. 

Regardless of the authenticity of the field, the term computer forensics is not very simple in answering the questions like how an electronic device was used in a crime. Was the computer a tool used to commit a crime? Or was the computer simply used to generate the digital landscape as a place for committing the crime.  Forensic expert must haul out the probative essentials and information from the target computers and then use this information to generate a reasonable scenario.

So, the multimedia forensics is the most exhilarating field in the world of cyber and computer forensics. Here in present article we will discuss one of the most appealing areas in world of forensic cyberspace i.e. multimedia forensics. Computer Forensics is the performance of collection, analysis, and report making on digital evidence to bring its value admissible in the court. Multimedia Forensics comes at a later stage, where scientific methods and techniques are used for  examination and analysis of the contents.

The specific topics covered in this article are:

What is Multimedia Forensics?

What are the approaches to Multimedia Authentication?

What are Digital Fingerprints?

What is Multimedia Forensics? 

Multimedia forensic is the application of investigating methodologies of forensics to the field  computer crimes. At the point when applied to the field of multimedia, advanced crime scene investigation in digital forensics began to confront difficulties, as multimedia is content that utilizes a blend of audio, video, pictures, and text. On account of the wide adoption of mobile devices, less expensive storage, and high transfer speed, the online clients are creating a huge amount of information. This development has pressed digital multimedia in the bleeding edge. The data amount is gigantic to the point that it has outperformed the capacities of the forensic specialists to successfully investigate and process the information. Multimedia forensics has now become a necessary piece of the Cyber Forensics. 

While this approach to digital safety sounds possible enough, it’s vital to recollect the sheer amount of information being handled during this scenario. In fact, the amount of data is therefore huge, it's surpassed the ability of forensic specialists to effectively processing and analyzing it. Plus, there are also the challenges of restricted time, various formats, short innovation cycles, and therefore the extremely dynamic environment. Tools are thus required to support the fact-finding method through exaggerated management, protection, interpretation, and visualization. The multimedia system community has developed some glorious solutions to the present downside, together with data extraction, automatic categorization, and compartmentalization. however there are still challenges and constraints involved, and most experts agree the time is correct to tailor and adapt multimedia forensics to fashionable demands.

This field of forensics includes the set of strategies utilized for the examination of multimedia signals like sound, video, pictures. It plans to:

Reveal the historical backdrop of digital content. 
Identifying the procurement gadget that created the data. 
Validating the uprightness of the content.
Retrieving data from multimedia signals. 

What are the ways to deal with Multimedia Authentication? 
It is well known fact that the data produced by multimedia can be altered or manipulated very easily. Though we cannot distinguish that manipulation. So whatever  we are seeing or listening to may have been altered spitefully for whatsoever reasons. Multimedia authentication can be simply defined as the basic step to confirming the authenticity  of the content produced by multimedia. By using different approaches of multimedia authentication, one can answers to the questions such as : (1) is the particular data originate from its alleged source? (2) has it been altered or manipulated  in either way? (3) if it has been changed then where and to what degree of manipulation was caused in that signal or data produce by multimedia.? 

Web content isn't just constrained to textual format; it arrives in a variety of assortments, so the criminological methodologies created to examine them should likewise fluctuate in scope. The objective here is to analyze the pictures, text, audio as well as video, so as to create a bit of coherent Forensic evidence

Multimedia Forensics is categorized into two broad groups: 
1. Active Image Authentication and 
2. Passive Image Authentication.
 
Active as well as passive image verification are additionally branched into different categories. These classes are discussed below in much detail:
1. Active Image Authentication: 
In this strategy, a known validation code is inserted in the image at the time of image formation or sent with the image for accessing its genuineness at the receiver end. Examination of this code confirms the originality of the image.

The active image authentication methodology is  sub-divided into two categories:

Digital Watermarking and Digital Signatures.

1.1. Digital Watermarking:
 In multimedia forensics, the watermarking is a method utilized to recognize the authentic source of a particular image or picture along with its user verification. This watermark is a specific code of  different characters which is entrenched in a digital text file or document file, audios, videos, images, or computer software program that offers the fine points to the investigating officer to observe more in relation to the images they came upon. This process is used to make it more easier for copyright holders to identify the individuals who tried to violate these rights. Moreover, a forensic watermark is able to alert an original user in the consequence that they unintentionally receiving an illegal document or software program. Furthermore, these  watermarks can be recurring at arbitrary locations in the content, which can make them fairly complicated to envisage and remove.

 In this process a computerized watermark which is entrenched into the picture either at the time of acquisition or during the stage of processing.
1.2. Digital Signatures: 
Digital signatures are another category of active authentication methodology. These signatures are similar to the handwritten signatures but they come out in an electronic form. The major role of  these signatures is to encode the contents of a multimedia file or document,  and consequently allowing those specific software applications to discover if there is any kind of alternation or tampering done in either ways. From a legal viewpoint, this furthermore allows the maker to show that a document existed at a certain date and time. 

Downsides of Active image authentication:
The authentication code needs to be inserted  in the image at the time of recording  by utilizing extraordinary tools therefore earlier information concerning the picture or image  becomes very important.
This method necessitate  a digital watermark or a computerized signature to be made accurately while recording the image, which limits its ability to handle specially equipped digital devices which constrains its capacity to deal with uniquely prepared computerized gadgets. 
As most of the pictures on the Internet, today don't have a watermark or a digital signature, which has made this process to mull over  some supplementary techniques –

2. Passive Image Authentication:
This method is also recognized as image forensics which requires only image itself for accessing its veracity without any secondary information for example a signature or watermark from the sender. This type of authentication based on the hypothesis that even if alteration in the image may not depart any visible trace but still they are expected to modify the primary statistics which shows that forgeries with the digital evidence may agitate the original properties as well as quality of the picture, even if there is no any visual clue has been found.

The passive image authentication are also subdivided in 2 categories
2.1. Forgery-type dependent -
This method is used to identify only some specific type of forgeries such as copy paste and splicing of the image which are relented on the nature and type of forgery carried out on the image. It is again sub-classified into two categories:

2.1.1. Copy-move detection: 
This is a well known photograph altering method due to the effortlessness with which it tends to be done. It includes replicating a few locales in the picture and moving it to some other region in the picture. Since the replicated area has a place on the same picture so the dynamic range and shading well matched with the picture. While detecting these kind of forgeries in the image some post-processing technique such as obscuring is utilized to diminish the impact of fringe abnormalities between the two pictures.    

2.1.2. Image-splicing detection:
This includes the assimilation of 2 or more photographs that significantly modify the original image to create a falsified image. Please note that when combining images with different backgrounds, it becomes difficult to make the borders indistinguishable. Detecting spliced images is a complex task involving the following methods:
• Composite regions are studied by various methods.
• The presence of abrupt changes between different areas that combine to create a composite image and their backgrounds provide valuable traces for detecting splicing in the image in question.

2.2. Forgery-type independent:
In these methods the detection of forgeries in the image is independent of forgery. Instead of forgery type this method is based on pieces of evidences which are left behind due to inconsistencies in illumination or lightening condition. 
It is also divided into two subcategories:

2.2.1. Retouching detection: 
Detection of retouching in the image is most frequently utilized for commercial as well as aesthetic applications. Retouching is habitually conceded out to increase or decrease the image features or to form a realistic masterpiece of 2 photographs which requires resizing, stretching or rotation of one of the photographs. The detection of retouching in the image is done by following methods:
Detection of the blurring, changes in the color and illumination in the forged image.
Detection of retouching in the questioned image becomes an easy task if the original one is available, otherwise the blind detection is a exigent task.

2.2.2. Lighting Conditions: 
The pictures which are combined during alteration are usually taken in dissimilar illumination. This illumination discrepancy in the composite photograph can be used for the identification of alteration in the image.

What are Digital Fingerprints?
The digital fingerprints are theoretically a coded form of binary digits which is created by statistical algorithms, on the basis of exceptionality these digital fingerprints can be compared with equivalent to the human fingerprint. In the world of multimedia, the digital fingerprint  tools can be used to recognize any piece of that media file such as a video or audio clip as being original itself, complete with its own unique features. This method includes the matching of digital fingerprint of the files which has been scanned  from different sites such as YouTube with that of copyrighted content to examine any kind of alteration if caused.

Although cryptographic tools and access control mechanisms provide secure delivery, but this protection is terminated as soon as the content is delivered to the end user. Digital fingerprinting has appeared to serve these post-delivery, identifying end users who are allowed to access the unencrypted text, but use it in unauthorized purposes.

In the process of digital fingerprinting, investigators track the illegal use of multimedia content using unique identifying information, known as “fingerprint,” which is embedded in the content before distribution. YouTube uses this technology to scan files and match the digital fingerprints they find with a database of copyrighted materials to determine if any intellectual property is being infringed.

Digital fingerprints are technically encoded strings of binary digits generated by mathematical algorithms; they are as unique as human analog fingerprints.
The more images and videos that continue to flood the Internet, the more difficult it becomes to protect information through forensic investigations. 

Conclusion
The more and more pictures, audios and videos keep on to deluge the internet for entertaining as well as educating the world at huge proportion, the more significant it becomes to safeguard it through forensic investigations. Nowadays  when online multimedia content  is  continuing to increase , it will become increasingly more essential for users and creators to comprehend the legal limitations of the virtual world and how they will be secured through caution and acquaintance.

References
Nampoothiri, V Parameswaran, and N. Sugitha. “Digital Image Forgery — A Threaten to Digital Forensics.” 2016 International Conference on Circuit, Power and Computing Technologies (ICCPCT), 2016, doi:10.1109/iccpct.2016.7530370.

Kumar, Shailesh, et al. “A Study on Authentication Framework by Using 2-D 3-D Image/Video Based Encryption.” International Journal of Engineering Research And, vol. V7, no. 02, 2018, doi:10.17577/ijertv7is020073.

Alamro, Loai, et al. “Figure 3: An Example of Image Splicing (A) and (B) The Genuine Images...” ResearchGate, 20 Nov. 2018, www.researchgate.net/figure/An-example-of-image-splicing-A-and-B-The-genuine-images-C-The-resulted-image_fig3_316667407.