Load management and security system through computer vision using raspberry pi

Project Title

Load management and security system through computer vision using raspberry pi

Project Area of Specialization Artificial IntelligenceProject Summary

First Part of Project: 

Life without electricity can never be imagined in this world, so it’s every human duty to save electricity.   And as now a days, world is adopting of automation of almost everything. Technology is trying to reduce the human effort and make the environment and electrical system efficient, reliable, intelligent and smart. There are so many automation systems installed in many organizations to save electricity like IR sensors are installed close to doors of any room to detect the presence of a person to turn on and turn off the lights automatically to reduce the human effort and to consume the light efficiently. And the world is trying to adopt the new systems which are more efficient, fast and reliable than the IR sensors we have now. 

Face detecting technique can be used to automate the electrical appliances like Lights. Fans, Air Conditioners, etc., and can result in saving electricity and avoiding energy losses. We will be taking the test case of a class room to automate the electric appliances using the face detecting technique. Raspberry Pi Module along with the Python language will be used to control the system, the algorithms will be developed to control the lights and the fans of the class rooms.

The system will work in a way that whenever someone enters the classroom, the CCTV camera will detect the entrance and the light and the fans will be switched on. The appliances will remain switched on until the camera detects the faces, the appliances will be switched off when everyone leaves the room and no detection occurs at the camera.

The system can be extended to an effective attendance system using face recognition techniques and can also be used for the security purpose at sensitive areas.

Second Part of Project:

Nowadays the number of thefts and identity fraud has become a serious issue. In order to avoid these thefts and identity fraud, a face recognition system must be established. The scope of this project is to develop a security access control application based on face recognition. The haar-like features is used for face detection and PCA algorithm is used for face recognition. In order to achieve a higher accuracy and effectiveness we use OpenCV libraries and python computer language. Training and identification is done in embedded device known as Raspberry Pi. During our paper we focus on accuracy increment by controlling parameters such as background, light and number of trainings. During our paper we also explicate cost issues of our application compared with commercial application.

The reason there is a demand to implement an access security system, is to control the people entering and going out of different buildings. As mentioned above, we need to be cautious about the privacy of the building. We have used a magnetic lock as a device to control the entrance of the building.

Project Objectives

Automation of electrical appliances in Classroom using Face Detection Method.To make existing automation system more robust, smart and fast.To reduce the electricity losses by smart and fast automation AI based system. And also to provide the suitable security systems for the entrance of person in buildings, so that the only authorized persons will be allowed to enter the building or in any room we want bto set up this system.

Project Implementation Method

Basic Steps:

Initially, we will design a prototype to test the algorithms, and the proper functioning of the automation system on small scale. The hardware for the project includes Raspberry Pi module, Camera, I/O Cables, Relays, SD Card.

The next step will be to get familiar with the algorithms of the Python language and its libraries. For that we will go through the computer tools like Computer Vision and OpenCV for effective implementation of face detection and then for face recognition.

First Method Implementaion :

Camera will be installed in class room for continuous monitoring, interfaced (connected) with programmed Raspberry Pi.  When students are present in the class, so that thier faces will be detected continously with the camera, as long as the faces will be detected or they are present in the class, the aplliances will be kept on. And once students will leave the class, camera will not find any face in the class so after some time the electrical appliances will be shut off and hence we can save electricity.

Second Part Implentation: 

We will have a micro SD card for Raspberry Pi Operating system and for data collection in which the images will be stored for the record of authorized persons.

And finally our final motive is to compare the image of the entering person with images stored in the micro SD card using face detection and face recognition techniques. And at the base of this comparison the Raspberry Pi will send a signal to relay by which any action can be taken like to open the door.

Benefits of the Project

First Part Benefits:

Efficient, reliable and robust automation of electricity will be achieved. We waste the electrical energy more then we use it so by this technology, electrical energy will almost be consumed not wasted.

Second Part Benefits:

Security thrait is now a days is big problem and to solve this issue, a reliable system is required for any organization. And the system provided by our project is reliable, robust and has a quick reponse. Apart from it, Implementation and installation is easy, low cost and it is reliable. 

Technical Details of Final Deliverable

The hardware for the project includes Raspberry Pi module for interfaing with the algorithm, Camera interfaced with raspberry Pi for monitoring, detecting and recognizing the faces, I/O Cables for the wiring between Rapberry Pi, Relays, camera and other things, Relays for switching the electrical appliances, SD Card for Raspberry Pi OS and database of authorized persons. Python language is there for our programming. Linux operating system will be used as Raspberry Pi uses debian. All the related libraries including openCV, numy and other libraries we will need.

Final Deliverable of the Project HW/SW integrated systemType of Industry Energy , Security 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)	8600
Raspberry Pi	Equipment	1	6500	6500
Pi Camera	Equipment	1	1100	1100
Door Lock	Equipment	1	650	650
Sd Card	Equipment	1	350	350