Model of an Intelligent Video Surveillance (IVS) System Prototype Using Deep Learning

Project Title

Model of an Intelligent Video Surveillance (IVS) System Prototype Using Deep Learning

Project Area of Specialization Artificial IntelligenceProject Summary

Surveillance refers to the close observation and monitoring of people's or objects' behaviour, activities, or other dynamic information in order to influence, manage, direct, or protect them. With technological advancements and the development of numerous intelligent prediction techniques, it is now possible to combine standard video surveillance with intelligence in order to recognize or make decisions based on scenarios. To efficiently prevent crimes and guarantee public security, intelligent video surveillance systems (IVS) based on image recognition are frequently used. A well-developed user-friendly and cost-effective product is not available for use because to the high complexity of processing real-time data and analyzing/understanding image contents.  In this project, we have developed a model of an Intelligent Video Surveillance (IVS) System Prototype using Deep Learning.

Project Objectives

1. To find the exact location of the image, the algorithm can check every portion of the continuous presence of an object in real-time videos of the intelligent video surveillance (IVS) system.

2. To design and implement the deep learning algorithm by using convolutional neural network techniques that can detect an object in real-time videos.

Project Implementation Method

Artificial Intelligence that deals with the theory behind artificial systems that extract information from images is known as "Computer Vision." Video Surveillance, on the other hand, is a word used to describe the use of videos to monitor any type of behaviour. It takes a human to keep an eye on the CCTV and a large amount of memory to record it. The massive amount of video storage and retrieval on demand is one of the most significant issues. Intelligent Video Surveillance System (IVS) is essential to avoid human resource depletion and to detect suspicious activities that pose a threat to safety and security.

Benefits of the Project

The experimental results compare current status with average test accuracy on real videos. Then this method provides test accuracy on real videos of the intelligent video surveillance (IVS) system. This intelligent video surveillance (IVS) system focuses on building a smart wireless security system that sends alerts to the owner by using the Internet in case of any trespass and raises an alarm optionally. Besides, the same set of sensors can also be utilized for home automation by making use of the same set of sensors. This research approach deals with the design and implementation of smart surveillance monitoring systems using sensors for devices. It increases the use of technology to provide essential security to our homes and for other control applications. This proposes an evaluation metric for crowd scene analysis methods. This metric estimates the difference between the calculated crowd count of the people and the actual crowd count in crowd scene videos.

Technical Details of Final Deliverable

The focus of this application of deep learning technique in detecting the exact count, involved persons, and the happened activity in a large crowd at all climate conditions. The project discussed the underlying deep learning implementation technology involved in various crowd video analysis methods. The role of video streams from CCTV cameras is equally important, and the algorithm for checking sensor data evolved from space using Deep Learning and using convolutional neural network techniques that can detect an object in real-time videos. Wireless CCTV cameras and an internet connection.

Final Deliverable of the Project HW/SW integrated systemCore Industry ITOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development GoalsRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	74590
camera	Equipment	1	40000	40000
security camera	Equipment	4	7000	28000
Fuel	Miscellaneous	8	250	2000
Meals	Miscellaneous	8	300	2400
internet packege	Miscellaneous	6	365	2190