Spatial video forgery detection

Project Title

Spatial video forgery detection

Project Area of Specialization Computer ScienceProject Summary SUMMARY

This project aims to develop a tool to detect spatial forgery detection in the videos based on a dataset composed of modified videos for forensic investigation. We will create a new data set of forged videos, primary purpose of developing and designing this new video library is for usage in video forensics, which can be consciously associated with reliable verification using dynamic and static camera recognition and developing the ground truth of the dataset. To the best of our knowledge, there exists no similar library among the research community. Footage from digital camcorders and smart phones will be included as well. Then we will use this data set to develop a tool for spatial forgery detection, it will take video as input in any format and as a output it will tell whether the video is forged or not, if it is then which technique is used to temper that video. Researchers will greatly benefit from such a library, as they will be provided with a general demonstration video to build a standard for their forensics algorithms.

Project Objectives OBJECTIVES

1. Develpment of Video Tampering Dataset (VTD) for researchers.
2. A tool will be developed to automatically detect the video is forged or not.
3. Tool will be able to process any type of video and will tell how many frames are forged in your video.
4. With this dataset, researchers are able to make a better assessment of the methods involved in their work.
5. Aimed at providing complete information and details regarding each type of video tampering found in the proposed dataset.

Project Implementation Method

The above developed dataset will be used to develop the tool that will detect the video forgery. Basically it will be based on Artificial intelligence, implementation steps are given below:

(1) Input

First or all we will give a video as a input in the tool and it be stored in the database for processing and future learning.

(2) Frame extraction

In the second step we will extract the frames of the video and store them on a specific location so that each frame will be processed

(3) Frame comparison

Now the extract frames will be compare based on some decision models(DOCFs) to find whether any of them frame is tampered on not 

(4) Generate output

Based on the above results the output will generate if any frame is found tampered then the video will be regard as tempered 

The whole process is shown in the following figure:-

Following tool will be used to develop the tool :-

1)  Matlab

2)  Python

3)  Adobe premiere Pro CC

4)  PyCharm(IDE)

Benefits of the Project BENIFITS

1. Proposed tool will accept all type of videos for forensic investigation.
2. Can be used to differentiate whether the video is original or edited.
3. It will differentiate which technique is used to tamper the video either spatial or tampered.
4. It will tell how many frames are forged and their locations in the video or an object is changed in the video.
5. This tool will help in all the current issues related to video forgery.
6. This tool will be helpful in the following fields:

• Media groups
• Insurance companies
• Marriage bureaus
• Investigation agencies
• Social media
• Courts of law

           

Technical Details of Final Deliverable Technical details:

1)  The tool will be able to take any type of video in any type of format

2)  It will separate all frame in the videos and examine individually 

3)  Support Vector Machine or other models will be used to detect whether the pair of consecutive frames is forged. If at least one pair of consecutive frames is detected as forged, the video segment is predicted as forged and the forged frames are localized.

Final Deliverable of the Project Software SystemCore Industry ITOther Industries IT Core Technology Artificial Intelligence(AI)Other Technologies Artificial Intelligence(AI)Sustainable Development Goals Quality EducationRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	78500
Nikkon D5200 camera	Equipment	1	38000	38000
IP cam	Equipment	1	5000	5000
Robotic camera	Equipment	1	5000	5000
Tripod	Equipment	1	3000	3000
Android Mobile	Equipment	1	18000	18000
Printing,Storage etc	Miscellaneous	1	4500	4500
Software Licence	Miscellaneous	1	5000	5000