csdf-mini-project
csdf-mini-project
On
by
Gaurav Boraste ( .)
Sahil Gaikwad ( .)
Ajay Vispute ( .)
Dr. B.S.Shirole
CERTIFICATE
of class B.E Computer; have successfully completed their mini project work on “Video
Forensic Analysis Tool’’ at Institute of Technology, Management & research, Nashik in the
partial fulfilment of the Graduate Degree course in T.E at the department of Computer
Engineering in the academic Year 2022-2023 Semester – I as prescribed by the Savitribai
Phule
Place : Nashik
Date :
ACKNOWLEDGEMENT
With deep sense of gratitude we would like to thanks all the people who have lit our
path with their kind guidance. We are very grateful to these intellectuals who did their
best to help during our project work.
We also thankful to our parents who providing their wishful support for our project
completion successfully and lastly we thanks to our all friends and the people who are
directly or indirectly related to our project work.
TABLE OF CONTENTS
1 CONTENT 4
2 ABSTRACT 5
3 INTRODUCTION 6
5 DESCRIPTION 8
6 PROJECT COMPONENTS 10
9 CODE 13
10 OUTPUT 16
11 CONCLUSION 18
1. ABSTRACT:
Video content has become an integral part of our digital world, but ensuring the authenticity
and integrity of video evidence is paramount for various fields, including law enforcement,
journalism, and digital forensics. The proliferation of video manipulation tools and deepfake
technology has heightened the need for sophisticated video forensic analysis tools. This
abstract introduces an advanced Video Forensic Analysis Tool designed to enhance the scrutiny
of digital video content, focusing on the verification of video authenticity and the detection of
tampering.
2. INTRODUCTION:
In an era marked by the ubiquity of digital video content, the authenticity and integrity of videos
have become subjects of paramount importance. From legal proceedings to journalistic
investigations, the need to ascertain the trustworthiness of video evidence is a crucial aspect of
our digital landscape. This introduction sets the stage for an exploration of the "Advanced
Video Forensic Analysis Tool," a cutting-edge solution designed to bolster the examination
and authentication of digital video content.
The proliferation of digital video technology has brought with it the power to document events,
tell stories, and provide evidence in ways previously unimaginable. However, this very power
has also given rise to an array of challenges. The ease with which videos can be edited,
manipulated, or fabricated has brought into question the veracity of video content, creating a
pressing need for advanced video forensics tools that can stand up to the task of preserving and
upholding the truth.
The Advanced Video Forensic Analysis Tool is a response to this evolving landscape. It was
conceived and developed as a comprehensive solution that goes beyond the basics of video
analysis, actively addressing the intricacies of video authenticity, and the detection of
tampering with a high degree of precision and accuracy.
This tool is a critical component in the toolkit of investigative professionals across diverse
sectors. From law enforcement agencies striving to secure justice through video evidence in
courtrooms to media organizations upholding journalistic ethics by verifying the credibility of
video sources, and digital forensics experts seeking to unravel complex cases, this advanced
solution serves as a versatile, all-encompassing resource.
The following sections of this document delve deeper into the functionalities, methodologies,
and features of the Advanced Video Forensic Analysis Tool. We will explore its capacity to
examine metadata, authenticate watermarks, conduct image and audio analysis, detect
deepfakes, and identify frame interpolation. Moreover, we will highlight its ability to generate
comprehensive reports with compelling visual representations of its findings, making it an
invaluable asset for professionals in the field.
The era of digital video is one of both promise and peril, and this tool represents a significant
step forward in safeguarding the authenticity of video content. It stands as a testament to our
commitment to ensuring that in a world where seeing is believing, what we see is, indeed, the
truth.
Software Requirements
Windows 7/8/10
Python 3
OpenCV library for video analysis
Hardware Requirements
Minimum I3 Processor
Minimum 2GB RAM
Minimum 128GB Hard Drive or SSD
4. DESCRIPTION:
The Video Forensic Analysis Tool represents a state-of-the-art software solution meticulously
engineered to serve as a foundational asset in the arsenal of investigators, legal professionals,
law enforcement agencies, digital forensics experts, and media organizations. In a
contemporary landscape where digital video content plays a pivotal role across multiple
sectors, ensuring the veracity and integrity of such content has never been more critical. This
tool emerges as a groundbreaking resource, addressing the complex challenges associated with
the authentication and validation of video evidence.
Comprehensive Metadata Examination:
One of the tool's foundational features is its capability to conduct a comprehensive examination
of video metadata. Metadata, which includes timestamps, geolocation data, device information,
and more, is often the key to establishing the authenticity and chain of custody of video
evidence. This functionality allows users to scrutinize these critical data points, ensuring that
the video's origin and history are consistent with its purported source.
Watermark Authentication:
Watermarks serve as a digital fingerprint, providing an unalterable mark of authenticity within
video content. This tool is equipped with robust watermark authentication capabilities,
allowing users to confirm the source and integrity of the video, making it an indispensable tool
for investigations and legal proceedings.
Advanced Image Forensics:
The tool is equipped with advanced image analysis capabilities that allow it to identify any
alterations within the video. These alterations might include image compositing, object
removal, or face manipulation. By flagging such potential manipulations or forgeries, this
feature significantly aids investigators in their pursuit of truth and justice.
Thorough Audio Analysis:
In the realm of digital video, audio plays a pivotal role in conveying information and context.
The Video Forensic Analysis Tool is adept at conducting in-depth audio analysis, detecting
signs of tampering, voice manipulation, audio splicing, and anomalies in background noise.
This level of audio verification is invaluable in preserving the fidelity of audio tracks within
video content.
Deepfake Detection:
In recent years, deepfake technology has presented a formidable challenge to video
authenticity. Utilizing machine learning models, this tool has the capacity to identify the
presence of deepfake elements within a video. By comparing facial and audio features to known
reference data, it can assist in identifying synthesized or manipulated content, further ensuring
the integrity of video evidence.
10
5. Project Components:
1. User Interface: Develop a user-friendly interface for users to interact with the tool.
2. Video File Input: Allow users to input video files (common formats like MP4, AVI, etc.).
3. Metadata Extraction: Extract metadata from the video file, including creation date,
camera model, and geolocation if available.
4. Hashing: Create a hash of the video file to ensure its integrity and detect any changes.
5. Frame Analysis: Implement a frame-by-frame analysis to identify any alterations, such as
splicing or frame insertion.
7. Report Generation: Generate a report with findings, including metadata, hashing results,
and any detected alterations.
8. User Documentation: Create user documentation on how to use the tool effectively.
Project Steps:
1. Install OpenCV:
2. Ensure you have OpenCV installed.
3. You can install it using pip: pip install opencv-python
4. Create the Python script:
11
12
9. Versatile Applications:
It is applicable across a wide array of professional sectors, including law enforcement,
journalism, digital forensics, and legal proceedings, ensuring its versatility and relevance in
different professional contexts.
10. Preserving Trust:
In a digital world where video manipulation is prevalent, the Video Forensic Analysis Tool
plays a pivotal role in preserving trust and integrity. It upholds the credibility of video evidence
and safeguards the principles of justice and ethical journalism.
Disadvantages
While the Video Forensic Analysis Tool offers a range of advantages, it's important to
acknowledge its limitations and potential disadvantages. These include:
1. Complexity:
The tool can be complex to use, requiring users to have a certain level of technical expertise in
video forensics and analysis. Novice users may find itchallenging to navigate and utilize all of
its features effectively.
2. Resource Intensive:
Video analysis, especially for high-resolution or lengthy videos, can be resource-intensive,
requiring powerful hardware and considerable processing time. This may limit its accessibility
for users with limited computational resources.
3. False Positives and Negatives:
Like any automated analysis tool, it is not infallible and may produce false positives and false
negatives in its assessments. Users need to exercise caution and rely on human judgment to
validate its findings.
4. Dependency on Data Quality:
The accuracy of its analyses is heavily dependent on the quality and integrity of the video data
being examined. Low-quality or heavily compressed video may yield less reliable results.
5. Constantly Evolving Manipulation Techniques:
The tool may not always keep pace with rapidly evolving video manipulation techniques. As
new methods and technologies emerge, it may require updates to maintain effectiveness.
6. Privacy Concerns:
When used in contexts where personal privacy is a concern, such as journalism, the tool may
inadvertently infringe on individuals' privacy by analyzing or extracting personal information
from video content.
13
7. CODE:
import cv2
import os
import hashlib
import json
# Function to extract video metadata
def extract_metadata(video_path):
metadata = {}
cap = cv2.VideoCapture(video_path)
metadata['frame_count'] = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
metadata['frame_width'] = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
metadata['frame_height'] = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
metadata['fps'] = int(cap.get(cv2.CAP_PROP_FPS))
cap.release()
return metadata
# Function to calculate video file hash (MD5)
def calculate_hash(video_path):
hash_md5 = hashlib.md5()
return hash_md5.hexdigest()
# Function to analyze frames for alterations
def analyze_frames(video_path):
14
cap = cv2.VideoCapture(video_path)
frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
altered_frames = []
prev_frame = None
for i in range(frame_count):
ret, frame = cap.read()
if ret:
if prev_frame is not None:
if not (frame == prev_frame).all():
altered_frames.append(i)
prev_frame = frame.copy()
cap.release()
return altered_frames
# Main function
if __name__ == '__main__':
video_path = 'sample_video.mp4' # Replace with the path to your video file
output_folder = 'output'
# Extract metadata
metadata = extract_metadata(video_path)
# Calculate hash
15
video_hash = calculate_hash(video_path)
# Create a report
report = {
'metadata': metadata,
'hash': video_hash,
'altered_frames': altered_frames
}
16
8. OUTPUT:
17
18
9. CONCLUSION:
This script will analyze the video file provided in the video_path variable. It will extract
metadata, calculate an MD5 hash of the video, and analyze frames for alterations, storing the
results in a JSON report