![Face Recognition Algorithm
Model Model.Train Model.Predict
Label index
&
Threshold
If Threshold
<= 800
Names[index]](https://image.slidesharecdn.com/frsslideson-170724094301/85/Face-Recognition-System-for-Door-Unlocking-22-320.jpg)
Face Recognition System for Door Unlocking
- 1. Group Members : Arslan Haider Zohaib Arshad Hassan Tariq Aamir Mehboob For Door Unlocking Raspberry Pi Based Face Recognition System Project Supervisor : Engr. Rizwan Qureshi
- 2. īĩ Why this project? âĸ Being a student of engineering, we have had experience in programming but never got a chance to develop on hardware such as the Raspberry Pi. The idea of working with real world hardware, and knowing that this technology is used worldwide in Blu-ray disks as well as by website like YouTube, Vimeo and iTunes Store, motivated me further to take up this project. īĩ Objectives of this project : âĸ Research into video processing âĸ Learn more about video streams âĸ Get familiar with Raspberry Pi âĸ Interfacing Raspberry Pi with Hardware. Project Objectives
- 3. âĸ A facial recognition system is a computer application capable of identifying or verifying a person from a digital image or video frame. âĸ It is typically used in security systems and can be compared to other biometrics such as fingerprint or eye iris recognition systems. Recently, it has also become popular as a commercial identification and marketing tool. Advantage of using face recognition: īĩ Least intrusive īĩ More Secure. What is Face Recognition System ?
- 4. âĸ There are many identification systems but face recognition is now-a- days more preferred. âĸ No Physical Interaction. âĸ Itâs not that expensive to install/implement. âĸ Like Security purposes (As of Door unlocking), Attendance system, face lock for mobile devices. âĸ Snapchat, Geo tagging(Facebook Auto Tags), law enforcement agencies etc. Why Face Recognition is Needed ?
- 5. âĸ An important difference with other biometric solutions is that faces can be capture from some distance away, with for example surveillance cameras. Therefore face recognition can be applied without the subject knowing that he is being observed (Security Purpose). Important Plus Point
- 6. âĸ Automatic Identification & Verification âĸ Database of faces âĸ Fisher Faces âĸ Comparison âĸ Face Match âĸ Applications Introduction
- 7. īĩ For face detection, OpenCV cascade classifiers will be used. īĩ These trained classifiers include detectors of face, eyes, nose and whole body, etc. Face Detection
- 8. âĸ The ViolaâJones object detection framework is the first object detection framework to provide competitive object detection rates in real-time. Haar Cascade Classifier
- 9. âĸ Eigen Faces âĸ Fisher Faces âĸ Local Binary Pattern Three Popular Algorithms For Recognition
- 10. âĸ Developed in 1997 by P.Belhumeur et al. âĸ Based on Fisherâs Linear Discriminant Analysis (LDA) âĸ Faster than eigenfaces, in some cases âĸ Has lower error rates âĸ Works well even if different illumination âĸ Works well even if different facial express. Fisher Faces
- 11. âĸ LDA maximizes the between- class scatter. âĸ LDA minimizes the within- class scatter. âĸ LDA seeks directions that are efficient for discrimination between the data. Class A Class B
- 12. âĸ Raspberry Pi Board âĸ Power Supply âĸ Relay âĸ Power Adapter âĸ Camera Module Components for Production of Project
- 13. âĸ Interfacing of camera module to capture live face image. âĸ Create a database of authorized person. âĸ Capture current face, save it and compare with database. âĸ Interface relay as output module. Proposed Work
- 14. Divided into 3 Parts: âĸ Camera Module âĸ Raspberry Pi Module âĸ Electronic bolt Lock Working and Methodology optional
- 15. Flow chart
- 16. 1st Phase
- 17. âĸ Take Frame âĸ Detect Face âĸ Extract the Face âĸ Resize âĸ Save Extracted Face Saving the Face Portion Take Frame Detect Face Extract the Face Resize Save Extracted Frame
- 18. Real Time Training Input Person Name Take Frame for 14 Sec Detect Face Save Face Loop terminated
- 19. âĸ Finally the Faces of persons will be saved in prove folder with the names of the persons. âĸ Those faces will then be used to recognize face. Cont.
- 20. 2nd Phase
- 21. âĸ Get Faces from Training Folder âĸ Compute Model âĸ Get Real-time Frame âĸ Detect Face âĸ Match that face âĸ Predict Name Real Time Face Recognition
- 22. Face Recognition Algorithm Model Model.Train Model.Predict Label index & Threshold If Threshold <= 800 Names[index]
- 23. Circuit Diagram
- 24. âĸ A Raspberry Pi is a general-purpose computer, usually with a Linux operating system, and the ability to run multiple programs. âĸ Very Low Cost($25-Rs 1550) for Model A & ($35-Rs 2200) for Model B/B+ âĸ Lighter, Smaller, Efficient. âĸ Lower power consumption (less than 5W). âĸ Supports Full HD video (1080p),Multiple USB Ports etc. âĸ Smartcard swapping, alcohol detection and agriculture humidity sensing etc. Why Raspberry Pi ?
- 25. Smart Surveillance Monitoring Security īĩ Living body detection and Spying Attendance System Criminal Recognition and Identification system Future Scope
- 26. īĩ The developed scheme is cheap ,fast, highly reliable and provides enough flexibility to suit the requirements of different systems. Extra Features: īĩ Counter: Counting the persons inside the specific room. īĩ Attendance system: Attendance and Displaying there names. Features
- 27. THANK YOU