(Affiliated to Guru Gobind Singh Indraprastha University, New Delhi)

A
Synopsis on
(MOTION DETECTION USING PYTHON)
Submitted in partial fulfilment of the requirements
for the award of the degree of
Bachelor of Computer Application

BATCH 2022-2025

Submitted by: Submitted to:

Student Name : JIYA MR. AZHAR AHMAD
Roll No: 01427402022 PROFFESOR ASSISTANT
 1. INTRODUCTION

Motion detection is a popular application in computer vision that involves identifying

and tracking changes or movements in a video or real-time footage. Using Python,
motion detection can be implemented with libraries such as OpenCV, which provides
tools for processing and analyzing video streams, and NumPy for numerical
computations.

Key Concepts in Motion Detection:

1. Background Subtraction: Identifying changes in a video by comparing each frame to
a static or dynamically updated background.
2. Frame Differencing: Comparing consecutive frames in a video to detect changes.
3. Contours and Thresholding: Highlighting the region of motion by thresholding and
finding contours.
4. Optical Flow: Tracking the motion of objects between frames using pixel movement
analysis.

Steps to Implement Motion Detection in Python:

1. Install Required Libraries:
o OpenCV: For video and image processing.
o NumPy: For matrix operations.
Install via pip:
bash
CopyEdit
pip install opencv-python opencv-python-headless numpy
2. Read and Process the Video Feed:
o Use OpenCV to capture video from a file or a webcam.
3. Perform Frame Differencing:
o Compare frames to detect changes.
4. Apply Thresholding and Find Contours:
o Highlight areas with motion and identify the moving objects.
5. Display Results:
o Show the original video and the detected motion in real-time.
.

2. OBJECTIVES AND SCOPE

The objective of this project is to develop a motion detection system using Python and
OpenCV that can identify and track movements in real-time or recorded video feeds.
By leveraging techniques such as frame differencing, background subtraction, and
contour detection, the system will efficiently highlight motion and provide visual
feedback for applications like security monitoring and automation This project is
applicable in various domains, including home security, robotics, wildlife
monitoring, and surveillance systems. It can process live or pre-recorded video,
detect motion in dynamic environments, and serve as a foundation for advanced
features like object recognition and alert systems.

3. Tools and Methodology

. Tools:

1. Python: Programming language for implementation.

2. OpenCV: For video processing and motion detection.

3. NumPy: For efficient frame manipulation.

4. Webcam/Camera: To capture live video.

5. IDE/Text Editor: Tools like PyCharm or VS Code for coding.

Methodology:

1. Capture Video: Use OpenCV to read frames from a webcam or file.

2. Preprocess Frames: Convert to grayscale and apply Gaussian blur to reduce

noise.

3. Background Subtraction: Identify motion by comparing frames with a

reference background.

4. Thresholding: Highlight motion by filtering out minor changes.

5. Contour Detection: Detect and outline moving objects.

6. Display Output: Show video with motion regions highlighted.

7. Testing: Validate under various conditions for accuracy.

 4. Plan of work
Plan of work would be as follows:
Sl. Months Work Schedule

1 0-1 Literature Study and Concept Development

2. 1-2 Feature extraction and data set collection

3. 2-3 Transfer of learning and Training the Model

4. 3-4 Testing the models at Testbeds and

Realbeds.

5.Significance
The significance of the motion detection system lies in its ability to
enhance security and automation across various applications. It plays a
crucial role in surveillance by enabling real-time monitoring of
environments such as homes, offices, and public spaces, detecting
unauthorized movements and potential threats. The system offers a cost-
effective alternative to expensive commercial solutions, leveraging Python
and OpenCV for efficient implementation. Its versatility extends to areas
like wildlife monitoring, robotics, and event-based video analysis.
Additionally, it serves as a foundation for integrating advanced
technologies such as object detection and machine learning, further
improving accuracy and functionality. Ultimately, the system provides
real-time insights, allowing users to respond promptly to detected motion,
making it an invaluable tool for security and automation.
 5. References:
[1] Neural network based age and gender classification for facial images by Thakshila
R. Kalansuriya and Anuja T. Dharmaratne, International Journal on Advances in ICT
for Emerging Regions 2014 07
[2] Deng, Jia, et al. "Imagenet: A large-scale hierarchical image database." Computer
Vision and Pattern Recognition, 2009. CVPR 2009. IEEE Conference on. IEEE, 2009.
[3] Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. "Imagenet classification
with deep convolutional neural networks." Advances in neural information processing
systems. 2012.
[4] Vedaldi, Andrea, and Karel Lenc. "MatConvNet-convolutional neural networks for
MATLAB." arXiv preprint arXiv:1412.4564 (2014).
[5] Zeiler, Matthew D., and Rob Fergus. "Visualizing and understanding
convolutional networks." Computer Vision-ECCV 2014. Springer International
Publishing, 2014. 818-833.
[6] Donahue, Jeff, et al. "Decaf: A deep convolutional activation feature for generic
visual recognition." arXiv preprint arXiv:1310.1531 (2013).
[7] Fahim Mannan, “Classification of Face Images Based on Gender using
Dimensionality Reduction Techniques and SVM” , 2007.
[8] Xiaofeng Wang, Mohd Ali2,3, Azliza Mohd Ali, Plamen Angelo, “Gender and Age
Classification of Human Faces for Automatic Detection of Anomalous Human
Behaviour”, 2017.
[9] Sebastian Lapuschkin, Alexander Binder, Klaus- Robert Muller, Wojciech Samek,
“Understanding and Comparing Deep Neural Networks for Age and Gender
Classification”, IEEE International Conference on Computer Vision Workshops ,
2017
[10] Sandeep Kumar, Sukhwinder Singh, Jagdish Kumar, “Gender Classification
Using Machine Learning with Multi-Feature Method”, 2019.
[11]http://www.vision.caltech.edu/Image_Datasets/
Caltech101/101_ObjectCategories.tar.gz
[12] Face Database :http://web.mit.edu/emeyers/www/face_databases.html
[13]Websites:https://www.forbes.com/sites/bernardmarr/2018/10/01/what-is-deep-
learning-ai-a-simple-guide-with-8-practical-examples/?sh=a78d138d4bad(Jan-2022).
[14] https://en.wikipedia.org/wiki/Facial_expression-(Jan-2022)
[15]https://developer.huawei.com/consumer-(jan-2022)
[16]https://www.thalesgroup.com/en/markets/digital-identity-and-security/
government/biometrics/facial-recognition