IVAS

Project Title

IVAS

Project Area of Specialization Artificial IntelligenceProject Summary

Nowadays, due to the increase in number of vehicles day by day, the associated security risks arises as well. Many security related issues can arise from parked vehicles, e.g. stolen vehicles, suspicious vehicles, Keeping track of vehicles entry/exit within a premises Currently, there is an absence of proper automated surveillance system for vehicles, e.g., there is no system which monitors which vehicle has entered or left and at which time.The above mentioned problem is related to the area of artificial intelligence and security. In artificial intelligence it is totally based on machine learning and computer vision. Computer vision is important because it is used to give better result of object recognition in real time and extracting feature more precisely.  Machine learning is important because it's possible to quickly and automatically produce models that can analyze bigger, more complex data and deliver faster, more accurate results – even on a very large scale. And by building precise models, an organization has a better chance of identifying profitable opportunities or avoiding unknown risks.An intelligent system which is envisioned to automatically monitor the environment or infrastructure with less or without human intervention. The main purpose of this object is to classify the vehicles on the basis of their features (dimension, color, size) i.e. whether it is car or bike etc. after the classification the system will give the count of vehicles which are present inside the parking or an organization. The number plate will be detected and time log of vehicle entry/exit will be maintained along with the number plate. The system would be intelligent enough to give alert about suspicious and overtime vehicle. It would also take the snap from the front of vehicle which then be stored in database for security concerns of any security agency.

Project Objectives

IVAS is developed to meet the following objectives.

make the datasets from the urban videos.

train the dartaset in order to classify the vehicles in two categories . i.e. Carike

on the basis of classification it will  it will give the count of the vehicles along the time stamp.

it will give the details at what time the car enter and exit the area.

it can be deployed in any parking enterance.

Project Implementation Method

IVAS project is base on Artificial Intelligence. The method we followed is to first make the dataset from urban videos. then train the dataset by using YOLO algorithm. there are many algorithms are there which are used for classification we have choose yolo in order to get result more precisely. we have use linux os for the whole implementation and then display the output in gui made through python language.

Benefits of the Project

it is cheap

cost and resource effective 

it can be deployed in any premises that have parking enterance.

requires only one camera and one user

easy to handle

easy to learn

reduce human effort

reduce physical check and balance

data can be used in investigations.

Technical Details of Final Deliverable

• Requirements Gathering deliverable is Problem statement
• Analysis Phase deliverables are Algorithms, equipment list.
• Collect Sample Dataset deliverables are Sample dataset which have been used in research paper “A comparative study of neural network and deep neural network based vehicle classification techniques”
• Train Sample Dataset deliverables are Features extraction through CNN/DNN algorithms
• Test Sample Dataset deliverables are Classify vehicles on the basis of their features and obtain accurate results
• Survey for CCTV camera deliverables are List containing cost and quality of different CCTV cameras
• Configure CCTV camera deliverables are Connect camera with the system
• Record videos deliverables are 4-5 recorded videos which is further used for creating dataset
• Create new Dataset deliverables are 1000-2000 images dataset of each vehicles, and 10-20 images for testing
• Train new Dataset deliverables are Features extraction through CNN/DNN algorithms
• Test new Dataset deliverables are Classify vehicles on the basis of their features and obtain accurate results
• Create website deliverables are Total count of vehicles, maintaining time log
• Integrate results with website deliverables are Connection between webpage and the classification results
• Display results on website deliverables are Display proper results Statistical display, Alerts, Live video.

Final Deliverable of the Project Hardware SystemType of Industry Transportation Technologies Artificial Intelligence(AI)Sustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	44500
GPU Gforce 1060 6gb	Equipment	1	37500	37500
power supply for gpu	Equipment	1	6000	6000
Prinitng cost	Miscellaneous	1	1000	1000
Travelling for survey	Miscellaneous	0	0	0