Design and development of Automated Attendance Monitoring System Using Face Recognition

Project Title

Design and development of Automated Attendance Monitoring System Using Face Recognition

Project Area of Specialization Augmented and Virtual RealityProject Summary

Face is the representation of one’s identity. So, we have prepared an automated student attendance system based on face recognition. This system is very useful in daily life applications especially in security and surveillance systems. The security systems on airport uses face recognition to identify suspects and the CBI (Central Bureau of Investigation) and FBI (Federal Bureau of Investigation) uses face recognition for criminal investigations. In our project also video framing is performed by accessing the camera through user friendly interface. The Face is detected and segmented from the video frame by using HOG (Histogram of Oriented Gradient) algorithm. In the first step or we can say in pre-processing stage, scaling of the size of the image is performed in order to prevent or reduce the loss of information. Then in next step, the ‘median filtering’ is applied to remove noise followed by the conversion of colour image into grayscale image. After that, CLAHE (Contrast Limited Adaptive Histogram Equation) is applied on the images to enhance the contrast of the image. Overall, we have created a program in python that take the image from the database and make all the necessary conversions for recognition and then verifies the image in the videos or in the real time by accessing the camera through user friendly interface. After the successful match is found then it marks the name and time of the person in attendance sheet. The human face is a unique representation of individual’sidentity. Thus, the face recognition is a type of biometric method through which identification of an individual is performed by comparing the real time captured image with the stored images of that person in the database. Currently, Facial Recognition System is widely spread due to its simplicity and fast performance. Some examples that represent implementation of this system are,the airport protection system and the FBI that uses face recognition system for criminal investigations by tracking suspects, missing children’s etc. On the other side, Facebook a popular social media site implements face recognition that allows the users to tag their friends in the photos for entertainment purpose. Apple allows users to unlock their mobile phones by face recognition.

Project Objectives

Currently manual student attendance marking technique is often facing a lot issues and a very slow process. Teacher’s or faculty calling names of student from their data sheet and student responding to them. But this existing process becomes very complex in large classes that consists so many students. Many times, students also mark proxies by responding to fake name. This makes disturbance in class and distracts the students during the exam times. Also, verifying the total students present by counting them after attendance, which takes a lot of time consuming. Apart from calling names attendance sheet is passed around classroom during lectures especially the classes consisting large number of students might find it hard to have attendance sheet being passed around the class. Douglas Ahlers, Bernie DiDario, Michael Dobson, in 2006 gave the concept of attendance tracking system. This framework consists of identity tags, with wireless communication capabilities, for each attende and the scanners for detecting the attende's tags as they enter in that allocated room. O.A. Idowu and O. Shoewu: Development of Attendance Management System by using Biometrics. Attendance is taken with the help of a finger print device and the recordsof attendance are stored in the database. Attendance is marked after successful identification. from Python or from the command line. It is a basic library constructed using dlib's cutting edge face recognition built with deep learning. The Dlib is a cross-stage open-source software library that is executed on various computing platform. The model has a precision of 99.38%. This provides a basic face recognition tool that allows you to perform face recognition on folder of pictures from the command line.

Project Implementation Method

In  the  proposed  system,  we  first  used  single  scale Retinex  with  bi-histogram  equalization  to  enhance  the quality  of  the  face  captured  then  later  applied  face recognition  algorithm  to  detect  and  recognize  the captured  face  by comparing  with faces  in  the database. Below  is the  brief explanation  of  Retinex enhancement algorithm  used  in  the  proposed  system.  Note  that,  the enhanced image produced  by  single scale  Retinex  is the input to the face recognition algorithm. Retinex  algorithm  is  one  of  the  image  enhancement algorithms  for  improving  the  quality  of  images  that provide color constancy and tonal  rendition..  The  general  model  of  Retinex algorithm is given by I(x,y)*L(x,y) Once image captured, single scale Retinex is applied to the image then take the bi-equalization, the output image is the enhanced image which can be use as an input to the face recognition algorithm. where ???? is the input image on the color channel, R is the Retinex output on the color channel and L is the normalized surrounded function. Single Scale Retinex This is one of image enhancement algorithms where by the output image is determined by the difference between the input image value and average of its neighborhood. We’d also use Histogram equalization it is the traditional technique for improving the contrast of the image but has drawback that can change the mean brightness of an image. Bi Histogram Equalization (BHE) is used for improving the histogram equalization (HE) method for contrast enhancement. BHE first finds average point in histogram of the image and then divides the histogram in to two segments based on this point. After that histogram equalization operation is applied on each segment. Single Scale Retinex with bi-histogram equalization provided a better result for image recognition algorithm. Integral image is a new representation which allow fast feature evaluation to determine whether the face is present or absent. AdaBoost The face recognition is of two forms: face location and face tracking. The face location locates the specific place which is suitable for detection. Face tracking works by tracking the length, breadth, size and pixels of the face as well as components such as eyebrows, eyes, nose, and mouth. Webcam was used as the input device for capturing the faces. Once the face is captured, single scale Retinex with bi-histogram equalization was used to enhance the sharpness of the image then face recognition algorithm is applied to the face captured. The algorithm can detect the face, extract features and recognize the face. The algorithm was implemented using MATLAB. Before the algorithm is used, the template images need to be stored in the database. Each person has 10 different faces patterns stored in the database. The database was designed to hold maximum number of 40 persons. Most of the face recognition algorithms have light variation problem. The images trained in dark room is very difficult to recognize in a bright room.

Benefits of the Project

Very few biometric technologies are attracting as much attention as facial recognition. Facial recognition technology offers a host of benefits such as in authentication, monitoring, access control, indexing, and maintenance of surveillance applications.Face recognition systems have led to the advancement of multimedia information access. Also, implementing network access control via face recognition not only makes it virtually impossible for hackers to steal a user’s password but also improves human-computer interaction. This is one of the reasons why facial recognition attendance systems are gaining popularity. Here we are going to explore the scope and advantages of the face recognition attendance system for companies. A facial recognition software captures and compares patterns on a person’s face and analyses the details to identify and verify the individual. While the underlying system is complex, the whole technology can be broken down into three steps. Face Detection: An essential step is locating human faces in real-time Transform Data: Once captured, the analogue facial information is transformed into a set of data or vectors based on a person’s facial features Face Match: The system matches the data above with the one in the database for verification Almost every big tech company including Amazon, Google, Microsoft, and Cisco is leading the effort to make face recognition more mainstream. There are many reasons to adopt the technology.
Entry and exit time monitoring done manually or with other biometric systems can be fully automated with facial recognition attendance systems. There is no need for human intervention or physical validation as the system’s advanced algorithms can locate and identify faces autonomously. It is effortless to track time for employees with facial recognition.
A facial recognisation attendance system can save business resources by automatic employee time tracking. Post pandemic there has been a significant increase in demand and adoption of contactless technologies. Face recognition attendance systems are not dependent on a few facial features but they are highly robust and identify a face on several data points. Therefore, these systems can screen for face masks and identify people without removing the mask or any change of facial attributes like beard, specs etc. It is a major advantage over any other biometric system as employees don’t have to take off their masks. With a face recognition attendance system, the entire environment is automated. You won’t just take the attendance but also automatically record the entry-exit time of the employees. It also adds to the security of the workplace as the system can recognize who left the designated area and when accurately.. Face recognition attendance systems are modern utilities that are a requirement of even the post-pandemic era. 

Technical Details of Final Deliverable

There are specific prerequisites for each platform that run applications based on the Face Verification. Minimum requirements that the clients must have in order to run thisprogram and acquire great outcomes are as follows: ? Hardware Specification: Processor: - 7 th generation i5.RAM: - Minimum 4 GB. Hard Disk: - Minimum 500 GB.Camera: - High quality. ? Software Specification: Platform: - Windows 8 or 10, LinuxLanguage Used: - Python Frontend tools: - PyCharm IDE, or Visual Studio Backend: - Database Directory, Attendance Excel Sheet.

Final Deliverable of the Project Hardware SystemCore Industry SecurityOther Industries Education , IT , Media , Telecommunication Core Technology Augmented & Virtual RealityOther Technologies Artificial Intelligence(AI)Sustainable Development Goals Decent Work and Economic Growth, Reduced Inequality, Partnerships to achieve the GoalRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	66500
Raspberry pi 4 Model B 8GB	Equipment	1	28500	28500
face detection camera	Equipment	1	18000	18000
wires and system	Equipment	1	10000	10000
added equipments	Miscellaneous	5	2000	10000