Real time vehicle detection, tracking and counting

Project Title

Real time vehicle detection, tracking and counting

Project Area of Specialization Artificial IntelligenceProject Summary

Exponential growth in population and increment in economic activities has caused gigantic growth of vehicles in all parts of the world. Due to this rapid growth countries are moving towards sustainable traffic management models
to solve the issue. Our proposed system will count number of cars in a particular area and it will also specify the type of vehicle. In this way this system  will help the authorities to make long lasting traffic models.

Detecting vehicles in a video stream is an object detection problem. An object detection problem
can be approached as either a classification problem or a regression problem. In the classification
approach, the image is divided into small patches, each of which will be run through a classifier
to determine whether there are objects in the patch. The bounding boxes will be assigned to
patches with positive classification results. In the regression approach, the whole image will be
run through a convolutional neural network directly to generate one or more bounding boxes for
objects in the images.

Project Objectives

The main goal of this project is to build a system that will track and keep the count of the vehicles
falling into the respective categories. We are building a system for real time detection and tracking of
vehicles to meet the security needs and also for the analysis purposes for the security agencies
providing the information about the intensity of specific traffic on a particular road what architecture
should be followed for future development keeping in mind the kind of traffic passing from certain
roads and what are the rush hours etc.
While there are a number of existing systems to give the key information about the traffic on the
roads to increase the security as well as other key information. We are trying to find an elegant
solution to this problem revealing as much as possible detail about the traffic on the roads within
a single system to reduce the effort and need of multiple systems.

Project Implementation Method

we will classify vehicles using deep learning and detection is perform by YOLO(YOU ONLY LOOK ONCE) we will train YOLO by ourself.And after that we will track vehicles using DEEP SORT Algorith and SORT algorith and check the accuracy and according to maximum accuracy we will choose algorith to deploy our system.

Benefits of the Project

through this we can maintain the data of number of vehicles using any road.

we can make development based on type of vehicle like if road is mostly used by LTV then it is better to build underpas rather then building bridge,

it can help law enforcment agencies to track the vehicle. and use road for VIP Protocol .

Technical Details of Final Deliverable

The main outcomes of this project are as follows:
1. A GUI that shows the vehicles passing by, categories of the vehicles and the number of vehicles displayed by a counter with their respective category giving a complete information about the traffic to prove as useful as possible for the security agencies in information gathering. We have adopted “less is more” approach to design this GUI.
2. A solid backend that handles app’s core functionality such as detecting the vehicles creating the bounding boxes around them tracking the vehicles and keeping the count
of the vehicles according to their respective category.

Final Deliverable of the Project Hardware SystemType of Industry IT Technologies Artificial Intelligence(AI)Sustainable Development Goals Decent Work and Economic GrowthRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	50014
camera	Equipment	1	50014	50014