Scribd
Upload
43 views3 pages

Synopsis Report

Uploaded by

01fe22mca004
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
43 views3 pages

Synopsis Report

Uploaded by

01fe22mca004
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Centre of Excellence in Visual Intelligence

NAME: - Vignesh V Shanbhag


USN: - 01FE22MCA004

Synopsis Report
Topic: Car Parking Space Counter Using Computer Vision in Machine
Learning
Aim: The demand for effective parking management systems has increased due to the
growing number of automobiles in metropolitan areas. Conventional approaches to parking
space management frequently lead to inefficiencies, traffic jams, and irritated users. The use
of computer vision with machine learning to construct a car parking space counter provides a
strong answer to these problems. The objective of this system is to optimize space use and
improve user experience by offering real-time parking spot monitoring and management.

Objectives:
1. Real-time Monitoring: Develop a system capable of real-time detection and counting of
available parking spaces.
2. Automation: Minimize human intervention in the parking management process.
3. Accuracy: Achieve high accuracy in detecting parked and vacant spaces under various
conditions.
4. Scalability: Ensure the system can be scaled to accommodate large parking areas.

Methodology
1. Data Collection:
- Image and Video Capturing: Collect images and videos from parking lots using high-resolution
cameras.
- Annotation: Manually label images to create a dataset for training the machine learning model.
2. Preprocessing:
- Image Enhancement: Apply techniques like noise reduction and contrast adjustment.
- Normalization: Standardize image sizes and formats for uniform processing.

3. Model Development:
- Object Detection: Use Convolutional Neural Networks (CNNs) to detect vehicles and parking
spaces.
- Training: Train the model on the labelled dataset using frameworks like TensorFlow or Py-Torch.
- Validation and Testing: Evaluate the model on a separate dataset to ensure its accuracy and
robustness.

4. Integration:
- Real-time Processing: Implement the trained model in a real-time system to process live video
feeds.
- User Interface: Develop a user-friendly interface displaying the current status of parking spaces.

Results
- Detection Accuracy: The system achieved an accuracy rate of over 95% in detecting available and
occupied parking spaces.
- Response Time: Real-time processing capabilities allowed for immediate updates on parking space
availability.
- Scalability: The system successfully managed large parking areas with minimal performance
degradation.

Challenges and Solutions


- Lighting Conditions: Variations in lighting were addressed using adaptive algorithms that adjust to
different light levels.
- Occlusions: Improved detection algorithms were developed to handle partial occlusions by other
vehicles or objects.

Future Work
- Advanced Analytics: Incorporate predictive analytics to forecast parking space availability.
- Integration with Smart City Infrastructure: Connect the system with other smart city
components for enhanced urban mobility.
- User Notifications: Develop a mobile application to notify users of available parking spaces in real-
time.

Conclusion
By giving precise, up-to-date information on parking space availability, the Car Parking Space
Counter employing Computer Vision and Machine Learning greatly enhances parking management.
In addition to improving user convenience, this technology helps improve urban traffic management.
Further developments in computer vision and machine learning will enhance and broaden this
system's capabilities.

You might also like