Attendance Automation using Face Detection & Recognition

Attendance Automation using Face Detection & Recognition

Alok Kr. Mishra, Nissar Ahmed, Durgesh Kumar, Aishwarya Yadav, Suraj Pal Singh

Krishna Engineering College

Abstract – While taking the online classes, it was easy to escape the classes for student. Also, it was not possible for a teacher to check the presence of the student, in the class during lectures. This problem was not only with regular online classes, it was also with the company who were giving online training. This issue can be extended for those job who were doing their work, from home. To overcome with this drawback, we can use face detection & recognition technique. In recent, there are multiple research work, that has been done related to it & latest technology are being develop, to overcome numerous problems in various field like bankcard identification, health monitoring, security and surveillance etc. This type of automation will help to complete the task easily, and will lead to an effective action taken in a short time. Hence, equipping us with a better technology to deal with the situation. The purpose of this work is, therefore, to provide a deep learning related to automation, attendance monitoring & face recognition. This will help us to develop study platform, and to track using an overhead perspective. The framework uses the object recognition paradigm to identify humans in video sequences. To increase the accuracy of the model, various self- learning algorithm is implemented. In this approach, the detection formula uses a pre-trained formula that's connected to an additional trained layer, that victimize associate degree to overhead human knowledge dataset.

Keywords Face recognition, eigen value, eigen faces, neural networks.

1. INTRODUCTION

   Face recognition is a very important analysis, that is used in various fields and disciplines in this modern era. This is often as a result of face recognition, & is used for multiple purpose like face identification for bankcard, ID card, security and closed-circuit television. It could be an elementary human behaviour, due to which it is essential for effective communication and interaction among people & help the system to analyse the surrounding[1].

   Progress has advanced to the point that face recognition systems are being demonstrated in real-world settings [2]. There is various factor that led in the growth of face recognition, such as : active development of algorithms, the availability of a large databases of facial images, and a method for evaluating the performance of face recognition algorithms. Within the literatures, face recognition drawback is often developed as: given static (still) or video pictures of a scene, determine or verify one or a lot of persons within the scene by comparison with faces keep in a very information.

   All face recognition algorithms consistent of two major parts:

   1. face detection and normalization and (2) face identification. Fully automatic algorithms are those that contain both the parts

      and those that consist of only the second part are called partially automatic algorithms. Partially automatic algorithms are given a facial image and the coordinates of the centre of the eyes. Fully automatic algorithms are only given facial images. On the other hand, the development of face recognition over the past years allows an organization into three types of recognition algorithms, namely frontal, profile, and view tolerant recognition, depending on the kind of images and the recognition algorithms. While frontal recognition certainly is the classical approach, view-tolerant algorithms usually perform recognition in a more sophisticated fashion by taking into consideration some of the underlying physics, geometry, and statistics.

      Profile schemes as stand-alone systems have a rather marginal significance for identification, (for more detail see [4]). However, they are very practical either for fast coarse pre- searches of large face database to reduce the computational load for a subsequent sophisticated algorithm, or as part of a hybrid recognition scheme. Such hybrid approaches have a special status among face recognition systems as they combine different recognition approaches in an either serial or parallel order to overcome the shortcoming of the individual components.

      Another way to categorize face recognition techniques is to consider whether they are based on models or exemplars. Models are used in [5] to compute the Quotient Image, and in [6] to derive their Active Appearance Model. These models capture class information (the class face), and provide strong values that deal with various variation in light, appearance etc. At the other extreme, exemplars may also be used for recognition.

      The ARENA method in [7] simply stores all training and matches each one against the task image. As far we can tell, current methods that employ models do not use exemplars, and vice versa. This is because these two approaches are by no means mutually exclusive. Recently, [8] proposed a way of combining models and exemplars for face recognition. In which, models are used to synthesize additional training images, which can then be used as exemplars in the learning stage of a face recognition system.

2. LITERATURE REVIEWS

   Face recognition systems have been grabbing high attention from commercial market point of view as well as pattern recognition field. It also stands high in researchers community. Face recognition have been fast growing,

   challenging and interesting area in real-time applications. A large number of face recognition algorithms have been developed from decades.

   This section gives an overview on the major human face recognition techniques that apply mostly to frontal faces, advantages and disadvantages of each method are also given. The methods considered are eigenfaces (eigenfeatures), neural networks, dynamic link architecture, hidden Markov model, geometrical feature matching, and template matching. There is substantial related work in multimodal biometrics. For example [13] used face and fingerprint in multimodal biometric identification, and [14] used face and voice. However, use of the face and ear in combination seems more relevant to surveillance applications.

   The attractiveness of using neural networks could be due to its non-linearity in the network. Hence, the feature extraction step may be more efficient than the linear Karhunen-Loève methods. One of the first artificial neural networks (ANN) techniques used for face recognition is a single layer adaptive network called WISARD which contains a separate network for each stored individual [15]. The way in constructing a neural network structure is crucial for successful recognition. It is very much dependent on the intended application. For face detection, multilayer perceptron [6] and convolutional neural network [7] have been applied.

   The classification time is less than 0.5 second, but the training time is as long as 4 hours. Reference [9] used probabilistic decision-based neural network (PDBNN) which inherited the modular structure from its predecessor, a decision based neural network (DBNN) [4]. The PDBNN can be applied effectively to 1) face detector: which finds the location of a human face in a cluttered image, 2) eye localizer: which determines the positions of both eyes in order to generate meaningful feature vectors, and 3) face recognizer. PDNN does not have a fully connected network topology. Instead, it divides the network into K subnets. Each subset is dedicated to recognize one person in the database. PDNN uses the Guassin activation function for its neurons, and the output of each face subnet is the weighted summation of the neuron outputs. In other words, the face subnet estimates the likelihood density using the popular mixture-of-Gaussian model.

   There is numerous research paper that define PCA (Principal Component Analysis) technique, for face recognition. PCA is a statistical approach used for reducing the number of variables in face recognition. In PCA, every image in the training set is represented as a linear combination of weighted eigenvectors called Eigen faces. These eigenvectors are obtained from covariance matrix of a training image set. The weights are found out after selecting a set of most relevant Eigen faces. Recognition is performed by projecting a test image onto the subspace spanned by the Eigen faces and then classification is done by measuring minimum Euclidean distance. A number of experiments were done to evaluate the performance of the face recognition system.

   The idea of using principal components to represent human faces was developed by Sirovich and Kirby and used by Turk and Pentland for face detection and recognition. The Eigen face approach is considered by many to be the first working facial recognition technology, and it served as the basis for one of the top commercial face recognition technology products. Since its initial development and publication, there have been many extensions to the original method and many new developments in automatic face recognition systems.

   Eigen faces is still considered as the baseline comparison method to demonstrate the minimum expected performance of such a system. Eigen faces are mostly used to:

   • Extract the relevant facial information, which may or may not be directly related to human intuition of face features such as the eyes, nose, and lips. One way to do so is to capture the statistical variation between face images.

   • Represent face images efficiently. To reduce the computation and space complexity, each face image can be represented using a small number of dimensions.

   The Eigen faces may be considered as a set of features which characterize the global variation among face images. Then each face image is approximated using a subset of the Eigen faces, those associated with the largest Eigen values. These features account for the most variance in the training set, of the face data that is register during recording.

3. METHODOLOGY

   A face detector has to tell whether an image of arbitrary size contains a human face and if so, where it is. One natural framework for considering this problem is that of binary classification, in which a classifier is constructed to minimize the misclassification risk. Since no objective distribution can describe the actual prior probability for a given image to have a face, the algorithm must minimize both the false negative and false positive rates in order to achieve an acceptable performance. This task requires an accurate numerical description of what sets human faces apart from other objects. It turns out that these characteristics can be extracted with a remarkable committee learning algorithm called AdaBoost, which relies on a committee of weak classifiers to form a strong one through a voting mechanism. A classifier is weak if, in general, it cannot meet a predefined classification target in error terms. An operational algorithm must also work with a reasonable computational budget. Techniques such as integral image and attention cascade make the Viola-Jones algorithm highly efficient: fed with a real time image sequence generated from a standard webcam, it performs well on a standard PC. The Viola-Jones algorithm uses Haar-like features, that is, a scalar product between the image and some Haar-like templates. More precisely, let I and P denote an image and a pattern, both of the same size N × N. The feature associated with pattern P of image I is defined by,

   I (I, J) 1P (i,j) is white I(I, J)1P(i,j) is black, 1 I N 1 j N1 I N 1 j N

   The sums of the pixels which lie within the White rectangles are subtracted from the sum of pixels in the grey rectangles. Two-rectangle features are shown in (A) and (B). Figure (C) shows a three-rectangle feature, and (D) a four-rectangle feature. To compensate the effect of different lighting conditions, all the images should be mean and variance normalized beforehand. Those images with variance lower than one, having little information of interest in the first place, are left out of consideration.

   Our face detection procedure classifies images based on the value of simple features. There are many motivations for using features rather than the pixels directly. The most common reason is that features can act to encode ad-hoc domain knowledge that is difficult to learn using a finite quantity of training data. For this system there is also a second critical motivation for features: the feature-based system operates much faster than a pixel-based system. More specifically, we use three kinds of features. The value of a two-rectangle feature is the difference between the sums of the pixels within two rectangular regions. The regions have the same size and shape and are horizontally or vertically adjacent. A three-rectangle feature – computes the sum within two outside rectangles subtracted from the sum in a centre rectangle. Finally, a four- rectangle feature computes the difference between diagonal pairs of rectangles. Given that the base resolution of the detector is 24 ×24, the exhaustive set of rectangle features is quite large, 160,000. Note that unlike the HMM basis, the set of rectangle features is over complete.

   In major, we have used the Viole-Jones Face detection to get the result. The Viola – Jones method for face object detection contains three techniques:

   o Integral Image for feature extraction, the Haar-like features is rectangular type that is obtained by integral image.

   Figure 1 – An Integral Image whose value will be calculated at point (x, y)

   Algorithm for Integral Image is as follow:

   1. Input: an image I of size N ×M.

   2. Output: its integral image II of the same size. 3. Set II(1, 1) = I(1, 1).

   1. for i= 1 to N do

   2. for j = 1 to M do

   6. II (i, j) = I(i, j) + II(i, j 1) + II(i1, j) II(i1, j 1)

   and II is defined to be zero

   1. Whenever its argument (i, j) ventures out of Is domain.

   2. End for

   3. End for.

   As shown in Figure, the value of the integral image at point (x,

   1. is the sum of all the pixels above and to the left.

      Figure 2 – Schematic Depiction of Detection Cascade

      • AdaBoost is a machine-learning method for face detection .The word – boosted means that the classifiers at every stage of the cascade are complex themselves and they are built out of basic classifiers using one of four boosting techniques (weighted voting).

      • Cascade Classifier is used to combine many features efficiently. The word cascade in the classifier name means that the resultant classifier consists of several simpler classifiers.

   Algorithm for building Cascade detector is as follow:

   1. User selects values for f, the maximum acceptable false positive rate per layer and d, the minimum acceptable detection rate per layer.

   2. User selects target overall false positive rate, Ftarget.

   3. P = set of positive examples.

   4. N = set of negative examples. 5. 5. F0 = 1.0; D0 = 1.0.

   6. I = 0.

   7. While Fi >Ftarget. 8. ii+ 1.

   9. ni= 0; Fi = Fi1.

   1. While Fi >f ×Fi1.

   2. nini+ 1.

   3. Use P and N to train a classifier with ni features using AdaBoost.

   4. Evaluate current cascaded classifier on validation set to determine Fi and Di.

   5. Decrease threshold for the ith classifier until the current cascaded classfier has a detection rate of at least d×Di1 (this also affects Fi).

   15. N .

   16. If Fi >Ftarget then evaluate the current cascaded

   detector on the set of non-face images and put any false detections into the set N.

4. WORK FLOW

   The below figures shown, is use to represent the various data flow diagram of the different process, running in the project. This will help us to understand the workflow & data transmit in the project modules. It also helps us to understand the working of a process. Data flow is important to get the better understanding of the model, with other models.

   DFD Level 0 Automated Attendance System:

   Figure 3 – DFD Level 0 Automated Attendance System

   The above figure tells that how the registration process of student will take place, when he/she will register on through web cam on the system.

   DFD Level 1 – Face Registration Module of Student:

   Figure 4 – DFD Level 1 Face Registration Module of Student

   The above figure will tell the registration process of the student, when he/she will register. The data will be added automatically in the database.

   DFD Level 2 – Face Recognition Module of Student:

   Figure 5 – DFD Level 1 Face Recognition Module of Student

   The above figure tells us about the recognition process done by the system. It will follow the same procedure for recognition of the face.

   The work flow of the project is explained by the following DFD:

   Figure 6 – Workflow of the project I

   First, we will register the user of various types like Admin, Organization, Student & Faculty. Once the registration is done, everyone will be able to login to Automated System. After that each one of the students have to upload a sort video to capture the face using face detection & recognition technique. Once the data of face will register, then the data will be saved in database of oracle.

   Figure 7 – Workflow of the project II

   When the registration of the organization, faculty, admin & student is done, then each one will able to join the class. After giving the access of webcam, continuous monitoring of student will start. If the presence is more than 75%, the attendance will be marked.

5. CONCLUSION

Our project is a platform that is created for purpose of study, kept a track on the student & presence in the class. Since it is based on automation, so it is easy to help to teachers, to get the remark. Its is not limited to class studies only, but also it can help in taking the online test without the presence of teacher & can be given an update in future.

In future, it is possible to keep track on the employee for the employers, since work from home is now becoming a trend now a days. Anywhere, where monitoring is required, our app can help an organization to keep their eye, without the human presence.

We have designed a true time machine-driven attending system, that reduces the time and resources that's needed, instead of taking attendance manually. This technique uses the technology of face detection and recognition. The system conjointly tells us whether the student is concentrating at school or not by checking the concentration of the person while attending the class. Various economical algorithms, measure the square & utilized, in order to obtained the specified result. This technique works well, within the ideal conditions and more improvement is created, when once the conditions don't seem to be ideal like correct illumination or lightning.

REFRENCES

[1] Ming-Hsuan Yang, David J. Kriegman, Narendra Ahuja, Detecting Faces in Images: A Survey, IEEE Transactions on Pattern Analysis and Machine Intelligence, Volume. 24, No. 1, 2002.

[2] Paul Viola and Michael J. Jones, Robust Real-Time Face Detection, International Journal of Computer Vision 57(2), p.p. 137154, 2004.

[3] William Robson Schwartz, Huimin Guo, Jonghyun Choi, Larry S. Davis, Face Identification Using Large Feature Sets, IEEE Transactions on Image Processing, Volume. 21, No. 4, 2012.

[4] Dayanand S. Shilwant, Dr. A.R. Karwankar, Student Monitoring by Face Recognition System ,International Journal of Electronics, Communication & Soft Computing, Science and Engineering, ISSN 2277-9477, Volume 2 ,2003.

[5] Matthew A. Turk and Alex P. Pentland, Face Recognition Using Eigen Faces, Computer Vision and Pattern Recognition, 1991. Proceedings CVPR91, IEEE Computer Society Conference, p.p. 586- 591, 1991.

[6] Yi-Qing Wang, An Analysis of the Viola-Jones Face Detection Algorithm, Image Processing on Line, Vol. 4, p.p. 128-148, 2014.

[7] Tarik Crnovrsanin, Yang Wang, Kwan-Liu Ma., Stimulating a Blink: Reduction of Eye Fatigue with Visual Stimulus, Conference on Human Factors in Computing Systems, p.p.2055-2064, 2014.

[8] Patrik Polatsek, Eye Blink Detection, Proceedings of 9th Student Research Conference in Informatics and Information Technologies, Bratislava, Slovakia, STU, 2013.

[9] Deepak Ghimire, Joonwhoan Lee, A Robust Face Detection Method Based on Skin Color and Edges, Journal of Information Processing System, Vol. 9, 2013.

[10] J. Kovac, P. Peer, F. Solina, Illumination Independent Color-Based Face Detection, IEEE Vol. 1,2

[11] P. Belhumeur, P. Hespanha, and D. Kriegman, Eigenfaces vs fisherfaces: Recognition using class specific linear projection, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp. 711720, 1997.

[12] M. Fleming and G. Cottrell, Categorization of faces using unsupervised feature extraction, In Proc. IEEE IJCNN International Joint Conference on Neural Networks, pp. 6570, 1990.

[13] B. Moghaddam, W. Wahid, and A. Pentland, Beyond eigenfaces: Probabilistic matching for face recognition, In Proc. IEEE International Conference on Automatic Face and Gesture Recognition, pp. 3035, 1998.

[14] A. Lanitis, C. Taylor, and T. Cootes, Automatic interpretation and coding of face images using flexible models, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp. 743 756, 1997.

[15] K. Jonsson, J. Matas, J. Kittler, and Y. Li, Learning support vectors for face verification and recognition, In Proc. IEEE International Conference on Automatic Face and Gesture Recognition, pp. 208 213, 2000.

[16] R. Brunelli and T. Poggio, Face recognition: Features versus templates, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, no. 10, pp. 10421052, 1993.