Search

In this paper we propose a novel transform domain steganography technique—hiding a message in components of linear combination of high order eigenfaces vectors. By high order we mean eigenvectors responsible for dimensions with low amount of overall image variance, which are usually related to high-frequency parameters of image (details). The study found that when the method was trained on large enough data sets, image quality was nearly unaffected by modification of some linear combination coefficients used as PCA-based features. The proposed method is only limited to facial images, but in the era of overwhelming influence of social media, hundreds of thousands of selfies uploaded every day to social networks do not arouse any suspicion as a potential steganography communication channel. From our best knowledge there is no description of any popular steganography method that utilizes eigenfaces image domain. Due to this fact we have performed extensive evaluation of our method using at least 200,000 facial images for training and robustness evaluation of proposed approach. The obtained results are very promising. What is more, our numerical comparison with other state-of-the-art algorithms proved that eigenfaces-based steganography is among most robust methods against compression attack. The proposed research can be reproduced because we use publicly accessible data set and our implementation can be downloaded.

Today's pandemic situation has transformed the way of educating a student. Education is undertaken remotely through online platforms. In addition to the way the online course contents and online teaching, it has also changed the way of assessments. In online education, monitoring the attendance of the students is very important as the presence of students is part of a good assessment for teaching and learning. Educational institutions have adopting online examination portals for the assessments of the students. These portals make use of face recognition techniques to monitor the activities of the students and identify the malpractice done by them. This is done by capturing the students' activities through a web camera and analyzing their gestures and postures. Image processing algorithms are widely used in the literature to perform face recognition. Despite the progress made to improve the performance of face detection systems, there are issues such as variations in human facial appearance like varying lighting condition, noise in face images, scale, pose etc., that blocks the progress to reach human level accuracy. The aim of this study is to increase the accuracy of the existing face recognition systems by making use of SVM and Eigenface algorithms. In this project, an approach similar to Eigenface is used for extracting facial features through facial vectors and the datasets are trained using Support Vector Machine (SVM) algorithm to perform face classification and detection. This ensures that the face recognition can be faster and be used for online exam monitoring.

Our current work proposes an experimental method that captures the thermal tomographic images under a heat flow process carried out for various boundary temperatures in a volumetric space. Thermal images generated thus experimentally, are then analysed using image processing technique for performance evaluation of heating quality (e.g. Uniform, Non-uniform etc.) In pre-processing stage, the images are converted into grey-scale images containing spatial pixel intensities. The pre-processed greyscale images are classified for quality of heating. This stage includes the normalisation, determination of Eigenface, training and testing. This Eigenface yields the feature value used for comparison with that of different classes like images for uniform and non-uniform heating etc. In Eigenface technique, the space of images is projected onto a low dimensional space using Principal Component Analysis. Eigenface is used to calculate pixel proximities between images. The likeness for the similarity of the test image as captured are found out after calculating the Euclidean Distance and PCA based Similarity Factor between the known Eigenfaces and test Eigenface.

Image Processing, which is a significant field of Machine Learning and falls under the adage of Artificial Intelligence, is used worldwide to achieve face detection and recognition. Image Processing focuses on two tasks, namely the enhancement of image for our interpretation and reading image data to store, transmit, and represent image perception via machines. The project focuses on the implementation of Machine Learning and Image Processing to automate attendance management in a corporate/personal space. In the current scenario, most corporate spaces make use of biometric systems to mark attendance, which is both time-consuming and costly. Setting up a working space with the biometric system also requires an individual with professional knowledge of the same. Our primary focus is to build a robust and automated security system capable of monitoring attendance without much human interference.

Multivariate Analysis (MVA) comprises a family of well-known methods for feature extraction which exploit correlations among input variables representing the data. One important property that is enjoyed by most such methods is uncorrelation among the extracted features. Recently, regularized versions of MVA methods have appeared in the literature, mainly with the goal to gain interpretability of the solution. In these cases, the solutions can no longer be obtained in a closed manner, and more complex optimization methods that rely on the iteration of two steps are frequently used. This paper recurs to an alternative approach to solve efficiently this iterative problem. The main novelty of this approach lies in preserving several properties of the original methods, most notably the uncorrelation of the extracted features. Under this framework, we propose a novel method that takes advantage of the,2,1 norm to perform variable selection during the feature extraction process. Experimental results over different problems corroborate the advantages of the proposed formulation in comparison to state of the art formulations. This work has been partly supported by MINECO projects TEC2013-48439-C4-1-R, TEC2014-52289-R and TEC2016-75161-C2-2-R, and Comunidad de Madrid projects PRICAM P2013/ICE-2933 and S2013/ICE-2933.

In today’s era, people upload their photographs on social media. Many applications like Facebook, Google, or in many fields like health care, advertising, criminal identification, etc., use facial recognition to identify the person or features of the object. Face tagging is an approach to find the person in the photograph automatically by using various techniques like LBP, regression analysis, eigenface, etc. This feature helps us to identify the person so that it becomes easier to get the details of every person tagged in the photograph. We can also find behavior or analyze their expression using the person’s face. Faces can be tagged without any contact with nature. This kind of image processing is better than image processing. It is more secure and cheap. In this paper, our objective is to find out efficiency in the future.

Face detection and recognition methods have enhanced a lot in the last decade in terms of accuracy, speed, and overall performance. This development has completely changed many systems. Attendance management is one example of this change. Earlier, the pen-paper-based method was used for marking and storing attendance. The improvement in biometric detection and recognition methods has resulted in the possibility to build a complete automated attendance management system. Face recognition based methods for attendance management has many advantages over the traditional attendance methods. The main challenge with the face recognition based method for attendance is to ensure that it recognizes every detected face. In this paper, previous works of several authors on face recognition based attendance management systems have been discussed, and a comparative study has been performed between the works. This paper shows that the need for cost-effective and network independent face recognition based attendance system having high accuracy still exists.

Attendance monitoring system is basic in all associations for checking the attendance of student or employee. In every association, attendance is taken physically by calling their register numbers or names and noted in their registers given by the association heads, and it consumes more time and effort. Face recognition is an example of recognition strategy and one of the most significant biometrics; it is utilized in an expansive range of utilizations. In our project, monitoring the attendance and classification of emotion is simplified by face recognition mechanism by using MATLAB and sending SMS to absentees by using GSM mechanism. Attendance is recorded in the proposed work with the aid of using a digital camera that is connected in the study room or working environment. This constantly shoots the pictures of student or employee and detects his/her face. It classifies the faces in pixels by PCA algorithm and examines the detected faces with the database and marks the attendance. Without physical connection of individual, it gives the participation of present student or employee in the association. For face recognition, the eigenface algorithm is utilized and classification of emotions is finished by Viola–Jones algorithm. In our work, when a face of an understudy matches with the dataset, it denotes the understudy as present and afterward at last it will send the SMS to the versatile which understudies is missing, via GPRS/GSM modem and Arduino. This system also detects the faces and classifies emotion whether interested or uninterested and loads the data in to the excel sheet.

Electroencephalography (EEG) signals allow to explore the functional activity of the brain cortex in a non-invasive way. However, the analysis of these signals is not straightforward due to the presence of different artifacts and the very low signal-to-noise ratio. Cross-Frequency Coupling (CFC) methods provide a way to extract information from EEG, related to the synchronization among frequency bands. CFC methods are usually applied in a local way, computing the interaction between phase and amplitude at the same electrode. In this work we show a method to compute Phase-Amplitude Coupling (PAC) features among electrodes to study the functional connectivity. Moreover, this has been applied jointly with Principal Component Analysis (PCA) to explore patterns related to Dyslexia in 7-years-old children. The developed methodology reveals the temporal evolution of PAC-based connectivity. Directions of greatest variance computed by PCA are called eigenPACs here, since they resemble the classical eigenfaces representation. The projection of PAC data onto the eigenPACs provide a set of features that has demonstrates their discriminative capability, specifically in the Beta-Gamma bands.

Today, face recognition systems play a crucial role in many access control and automatic identification systems. However, these systems still have shortcomings that reduce their performance efficiency. In this paper, a novel face recognition framework is introduced, combining the Eigenfaces algorithm and image registration. Firstly, the collected face images are preprocessed, then the Eigenfaces algorithm is applied to them for obtaining the reference eigenvectors. After that, three test images are captured using a webcam, and the images' faces are detected using the Viola-jones algorithm. The detected faces are registered to the collected face images, and the detected face with the lowest mean square error is selected for subsequent steps. Next, the selected detected face's eigenvector and the distance between it and reference eigenvectors are calculated, respectively. The minimum distance is then compared with a manual threshold to recognize the person as an unknown or known person. If the person is recognized as a known person, the person's identity is identified as the person belongs to the minimum distance. For validating the presented method, a public and an exclusive face image database are used. The obtained results indicate that the proposed framework achieved a better performance than traditional similarity-based methods to recognize known and unknown persons and identify known persons.

Face Recognition is most used for biometric user authentication that identifies a user based on his or her facial features. The system is in high demand, as it is used by many businesses and employed in many devices such as smartphones and surveillance cameras. However, one frequent problem that is still observed in this user-verification method is its accuracy rate. Numerous approaches and algorithms have been experimented to improve the stated flaw of the system. This research develops one such algorithm that utilizes a combination of two different approaches. Using the concepts from Linear Algebra and computational geometry, the research examines the integration of Principal Component Analysis with Delaunay Triangulation; the method triangulates a set of face landmark points and obtains eigenfaces of the provided images. It compares the algorithm with traditional PCA and discusses the inclusion of different face landmark points to deliver an effective recognition rate.

In this paper, the facial expression recognition system is described based on Eigenface. Facial expression recognition has become increasingly common in several applications in the fields of video conferences, surveillance, financial services, health treatment, and access control system. The performance of recognition depends mainly on several challenges. Lighting variations are also the most challenging in the recognition process. In this system, illumination invariant techniques applied for decreasing the lighting affected faces as a preprocessing stage. Eigen faces are used as feature extraction step. K-NN and Multi-SVMs classifier are used to identify the facial expression and to test the performance of the proposed system. The standard dataset of JAFFE is used in the experiments and 84.29% accuracy is achieved for facial expression recognition. The accuracy of the framework is contrasted with the state of the art methods tested on the JAFFE database.

Face recognition (FR) in an unconstrained environment, such as low light, illumination variations, and bad weather is very challenging and still needs intensive further study. Previously, numerous experiments on FR in an unconstrained environment have been assessed using Eigenface, Fisherface, and Local binary pattern histogram (LBPH) algorithms. The result indicates that LBPH FR is the optimal one compared to others due to its robustness in various lighting conditions. However, no specific experiment has been conducted to identify the best setting of four parameters of LBPH, radius, neighbors, grid, and the threshold value, for FR techniques in terms of accuracy and computation time. Additionally, the overall performance of LBPH in the unconstrained environments are usually underestimated. Therefore, in this work, an in-depth experiment is carried out to evaluate the four LBPH parameters using two face datasets: Lamar University data base (LUDB) and 5_celebrity dataset, and a novel Bilateral Median Convolution-Local binary pattern histogram (BMC-LBPH) method was proposed and examined in real-time in rainy weather using an unmanned aerial vehicle (UAV) incorporates with 4 vision sensors. The experimental results showed that the proposed BMC-LBPH FR techniques outperformed the traditional LBPH methods by achieving the accuracy of 65%, 98%, and 78% in 5_celebrity dataset, LU dataset, and rainy weather, respectively. Ultimately, the proposed method provides a promising solution for facial recognition using UAV .

The eigenface algorithm is a collection of eigenvectors used for face recognition through computers. The face recognition system is part of image processing that recognizes faces based on imagery that is stamped and stored in an image file in JPEG format. Face recognition problems can be solved through the implementation of an algorithm. The algorithm used in this study is the eigenface algorithm. The input image is stamped through a laptop camera with a size of 320 pixels x 240 pixels and reduced to 100 pixels x 100 pixels to be saved as a master file of the face with various facial expressions is a forward-facing position without a smile, facing forward a thin smile, facing forward with a big smile, head tilted to the left, and head tilted to the right. The purpose of this research is to build face recognition software using eigenface algorithms. The results showed that faces could be recognized using the eigenface algorithm with an average accuracy rate of 85%.

In the vast field of the artificial intelligence, face recognition is the technology that progresses the fastest. Identifying a human face in images or videos is a time-limited process that requires huge computing capacity. Modern GPUs offer massive computing power that could be efficiently applied in many applications. In this paper, we present a novel implementation of the Eigenface PCA face recognition algorithm, based on several optimization technics. Using NVDIA GeForce 750M GPU, the experimental results showed that our implementation outperformed the existing algorithms by 3x factor.

Convolutional Neural system arrangement for anticipating the likelihood of relationship of a person with others in a social affair is the main objective of this paper. The framework joins the convolutional neural system and AI calculations. The coordinated applications are face identification, face bunching and face identifier. This technique is fit for quick order and gives preferred execution over the eigenfaces approach. The DLIB calculation encodes the face into 128D exhibit and stores it in record taking care of framework. The framework accepts video as source and employments face acknowledgment Chinese murmur bunching calculation to group the essences of a person. In view of the information from the bunch results, the apriori/affiliation rule mining calculation finds the affiliated likelihood. The framework is coordinated with google cloud to give application-based interface which makes application programming interface calls. Association rule mining algorithm is used to find the confidence probabilities of a person with other individuals of the video clipping.

The current activity of drug inspection is usually carried out at school or university. This procedure, however, is less effective and efficient, as the urine samples are taken randomly. In many cases, the suspect student is not present or escapes the urine or hair inspection. A predictive drug user is needed, where only students suspected of positive drug use are selected for a urine test. To handle this problem, we need a system to predict suspect positive drugs. The dataset is generated from online sources by collecting and pre-processing 30 images of people before and after drug. We compare two algorithm local binary pattern and Eigenfaces for predicting suspect positive drugs based on face images. The experiment shows that the result of the prediction using Local binary pattern is better than the prediction using Eigenfaces. However, a higher accuracy of prediction reaches only 75 %.

— Computer visions and their applications have become important in contemporary life. Hence, researches on facial and object recognition have become increasingly important both from academicians and practitioners. Smart gadgets such as smartphones are nowadays capable of high processing power, memory capacity, along with high resolutions camera. Furthermore, the connectivity bandwidth and the speed of the interaction have significantly impacted the popularity of mobile object recognition applications. These developments in addition to computer vision’s algorithms advancement have transferred object’s recognitions from desktop environments to the mobile world. The aim of this paper to reveal the efficiency and accuracy of the existing open-source facial recognition algorithms in real-life settings. We use the following popular open-source algorithms for efficiency evaluations: Eigenfaces, Fisherfaces, Local Binary Pattern Histogram, the deep convolutional neural network algorithm, and OpenFace. The evaluations of the test cases indicate that among the compared facial recognition algorithms the OpenFace algorithm has the highest accuracy to identify faces. The findings of this study help the practitioner on their decision of the algorithm selections and the academician on how to improve the accuracy of the current algorithms even further.

The recognition of human faces plays an important role in many applications, for example in video surveillance and the management of facial image databases. This paper will design and implement a security system based on a machine learning algorithms. Principal Component Analysis (PCA) is the algorithm that represents the faces economically. It extracts the most dominant Eigenfaces from the present set of the faces. Comparison of video frames can be done by using this technique. Faces can be recognized in frames using the haar cascade to extract the characteristics of a human face. The SVM algorithm is used to classify between data sets using the kernel. The performance of the identification system also depends on the extraction of the attributes and their classification in order to obtain accurate results. These algorithms give different accuracy rates under different conditions, as observed experimentally. The precision and efficiency with which the model identifies people is the real added value of this paper.

The authenticated door-lock security system for smart-home is a critical Internet-of-Things (IoT) application. The most pressing bottleneck of running algorithms in IoT devices is the availability of low computation resources. The state-of-the-art algorithms which work significantly well in high-end computing unit more often fail to operate in IoT devices like Raspberry pi. This work targets design of a face authenticated door-lock security system in Raspberry pi and, therefore, evaluates the performance of conventional face detection and recognition models. The local binary pattern histogram (LBPH), eigenface and fisherface algorithms are considered for evaluation. The evaluation analyzes the accuracy of a model in detecting human faces, its computation time and vulnerability to photo hacking while implemented in Raspberry pi platform. It covers a large number of stored faces and the cases of variation in angle and illumination in the face image. The python and c++ bindings to the opencv framework are considered to create a fully functional face recognition system hardware.

The attendance system is used to track and monitor whether a student attends a class. There are different types of attendance systems like Biometric-based, Radiofrequency card-based, face recognition based and old paper-based attendance system. Out of them all, a Face recognition based attendance system is more secure and time-saving. There are several research papers focusing on only the recognition rate of students. This research focusing on a face recognition based attendance system with getting a less false-positive rate using a threshold to confidence i.e. euclidean distance value while detecting unknown persons and save their images. Compare to other euclidean distance-based algorithms like Eigenfaces and Fisherfaces, Local Binary Pattern Histogram (LBPH) algorithm is better [11]. We used Haar cascade for face detection because of their robustness and LBPH algorithm for face recognition. It is robust against monotonic grayscale transformations. Scenarios such as face recognition rate, false-positive rate for that and false-positive rate with and without using a threshold in detecting unknown persons are considered to evaluate our system. We got face recognition rate of students is 77% and its false-positive rate is 28%. This system is recognizing students even when students are wearing glasses or grown a beard. Face Recognition of unknown persons is nearly 60% for both with and without applying threshold value. Its false-positive rate is 14% and 30% with and without applying threshold respectively.

Most face recognition algorithms are easily affected by external factors, or require the recognizer to stand in a fixed position, so the recognition is slow and the accuracy is not very high. This paper implements an application platform of face recognition based on Open CV. Firstly, the face image is collected and preprocessed by optimizing the algorithms of Eigenfaces, Fisherfaces and LBP platform, which are used for face recognition by cooperation with each other. In addition to the restrictions on the recognition of the target to be detected by the face recognition system, eye location is added to the traditional AdaBoost based face detection algorithm to realize the function of face module training and face recognition. After testing, the system is proved to have a friendly interface, stable operation, good robustness to face location and environmental light changes, and it can detect and recognize face in real time quickly and accurately.

The face identification system is one of the most emerging methods for authentication of user; it is drawing wide attraction to the surveillance system which reflects innovation in a video surveillance system. This system here represents the automated attendance system using real-time computer vision algorithms and adaptive techniques to track the faces during a specific period of time. Our system works on eigenface recognizers and Intel’s Haar cascades which make the attendance-taking process easier and less time-consuming rather than the traditional process. Our system provides the cheapest solution rather than a previous biometric system like fingerprint authentications. The recorded data is being compared with the training dataset, and the attendance is recorded if the match is found with the help of Python libraries. The camera is being installed at the entry location, so that attendance is recorded as soon as the match of the person entering the particular area is found. So, our main aim is to provide an alternative which is very much convenient to process the attendance and also is very much safe and authentic to have faced as a security option.

Nowadays robots are one of the most important technologies and face recognition is crucial for human-robot interaction. Face recognition has direct benefits in the commercial and law enforcement fields and it enables robots to perform a wide variety of roles such as assistance, search and rescue, military and so on. In this paper we propose a Cloud-Edge architecture for robots that enables the use case of face recognition in deadline constrained environments. We formulate a mathematical model for a Data Capsule which represents structured units of data in a time series. We design the components on each layer of the architecture. We propose a deadline aware scheduler in the Fog that acts as a proxy for the processing platforms in the Cloud and we design two face recognition applications, one in the Edge for robots that is implemented with eigenfaces and one in the Cloud for the processing platforms with deep neural networks (DNN). We evaluate the performance of the face recognition applications by running a workload that consists of a well-known labelled image data set. We test the ability of the Fog scheduler to launch jobs on time when strict deadlines are in place and it runs a heavy workload of jobs.

Fast and accurate user identification and verification is always desirable. Face recognition, which is machine recognization of person face by analysing patterns on facial features is becoming important for security and validation. Less interaction from user contributes high enrolment as well as easily applicable for current technology further adds its importance. In this regard, we propose a Convolutional Neural Network (CNN) based face recognition technique previously done with eigenfaces[8] but CNN has better accuracy. Entire process is divided into four phases: capturing the image, features extraction, classification and matching. At&t faces dataset is used in this paper. Input images are first fed for face detection. Face detection in input images are performed using Viola Jones algorithm. Convolutional Neural Network (CNN) is applied for feature extraction and classification. The result obtained in this paper shows that - recall is 0.992, precision is 99.4, f1 score is 99.1 and f1 beta score is 0.992 for 70-30 split of dataset, i.e, 70% dataset used as training dataset and 30% as testing dataset.

While recognizing any individual, the most important attribute is face. It serves as an individual identity of everyone and therefore face recognition helps in authenticating any person's identity using his personal characteristics. The whole procedure for authenticating any face data is sub-divided into two phases, in the first phase, the face detection is done quickly except for those cases in which the object is placed quite far, followed by this the second phase is initiated in which the face is recognized as an individual. Then the whole process is repeated thereby helping in developing a face recognition model which is considered to be one of the most extremely deliberated biometric technology. Basically, there are two type of techniques that are currently being followed in face recognition pattern that is, the Eigenface method and the Fisherface method. The Eigenface method basically make use of the PCA (Principal Component Analysis) to minimize the face dimensional space of the facial features. The area of concern of this paper is using the digital image processing to develop a face recognition system.

El objetivo fue la implementación de dos algoritmos de reconocimiento facial para la identificación de personas, bajo condiciones reales en este caso un domicilio, se examinó las investigaciones relevantes en la identificación facial como los métodos de Análisis de Componentes Principales (PCA) y el Análisis Lineal Discriminante (LDA), procesos matemáticos utilizadas por los algoritmos Eigenfaces y Fisherfaces, que son técnicas de aproximación estadística para clasificar datos en este caso rostros, con mínima complejidad computacional. El sistema de identificación de personas propuesto, desarrolló algoritmos en Matlab, describiendo sus etapas principales de procesamiento como la adquisición de imágenes, pre procesamiento, extracción de características faciales, entrenamiento o clasificador, éstos algoritmos se sometieron a evaluación de eficacia en la identificación de personas, utilizando una base de rostros públicas de AT&T conocida como ORL Database of Faces, con 400 imágenes de rostros de distintas personas con variadas expresiones, adicionalmente se elaboró un base de rostros domiciliar, con 48 imágenes para el entrenamiento del sistema. Como resultado de la evaluación bajo condiciones de laboratorio se obtuvo una eficiencia superior al 90%, mientras que, al ser evaluados en condiciones reales, los algoritmos presentan eficiencias de identificación que oscilan del 50 al 80%, factores que dependen de la distancia, cámara, calidad de imagen, e iluminación. Estableciéndose las condiciones mínimas de requerimientos para la implementación de un sistema de vigilancia domiciliar, con porcentajes de efectividad favorables en estos ambientes, proponiendo una innovación en el uso de la tecnología e ingeniería actual respecto a video vigilancia. The objective of the project is the implementation of two facial recognition algorithms for the identification of people, under real conditions, in this case, an address. Relevant investigations in facial identification have been explored, such as Principal Component Analysis (PCA) and Discriminant Linear Analysis (LDA) methods, mathematical processes used by the Eigenfaces and Fisherfaces algorithms, which are statistical approximation techniques for classifying data, in this case, faces, with minimum computational complexity. The proposed people identification system, developed its algorithms in Matlab, describing its main processing stages such as image acquisition, pre-processing, extractions of facial features, training or classifier, these algorithms will be evaluated of efficiency in the identification of people, used for this a base of public faces of AT & T known as ORL Database of Faces, with 400 images of different faces forgive with various expressions, in addition to this a base of faces domiciliary, with 48 images will be elaborated for system training. As a result of the evaluations under laboratory conditions they present an efficiency of over 90%, while when evaluated in real conditions, the algorithms show efficiencies ranging from 50 to 80% depending on the distances between the subject and the camera, as well as it also depends on the image quality that can be acquired with them. In this way, minimum requirements have been established for the implementation of a home surveillance system, in which favourable percentages of effectiveness will be obtained in these environments, proposing an innovation in the use of current technology in relation to video surveillance. Palabras claves: PCA, Eigenfaces, LDA, Fisherfaces, Base de Datos Públicas, Algoritmo, Reconocimiento. Keywords: PCA, Eigenfaces, LDA, Fisherfaces, Public Database, Algorithm, Recognition.

The process of preventing unauthorized use of computers is called computer security. A convenient user authentication device is more essential than ever before. The major intend of this paper is to provide a suitable and reliable authentication method using keystroke and mouse dynamics combined with face recognition. Using user-id and password is considered as the most basic type of authentication. An unauthorized person can access a genuine user’s resources by exploiting an unreached unlocked computer or possibly by stealing the password of user. Without continuous authentication it is not possible to say who the real person is. So the paper is to provide a combination of static and continuous authentication. This is a biometric system to avoid impersonation and is a hybrid security system. Typing pattern in the keyboard of a user is combined with mouse interaction behavior to perform continuous analysis of a user and extracting the face features to perform static authentication. Dataset was collected and analysis was done continuously and transparently during the whole session. The keyboard, mouse interaction behaviors, and face features are unique to each individual. The method provides a probable solution to authenticate rightful users statistically and endlessly.

Face recognition is an important concept, which has generally considered in the course of recent decades. Generally image location can be considered as an extraordinary sort of item recognition in PC vision. In this paper, we explore one of the vital and very effective systems for conventional item discovery using convolutional neural network (CNN) method, that is, a gentle classifier development for resolving the substance identification problem. That in recognizing the face of images as the problem is very difficult one, and so far no quality results are been obtained. Usually, this problem splits into distinctive sub-issues, to make it simpler to work predominantly identification of face of a picture pursued by the face acknowledgment itself. There are several tasks to perform in between such as partial image face detection or extracting more features from them. Many years there are numerous calculations and systems have been utilized such as eigenfaces or active shape model, principal component analysis (PCA), K-nearest neighbour (KNN), and local binary pattern histograms (LBPH), but accurate results have not been identified. However because of the drawbacks of previously mentioned techniques in my study, I want to use CNN in deep learning to obtain best results.

Biometrics are used to characterize a person's DNA, hand geometry, confront, and so on or behavioral qualities, for example, hand signature, voice tone, keystrokes et cetera. For that reason, these organic attributes are remarkable for each person. Much of the time, confront acknowledgment related advancements are winding up more mainstream among biometric-based advances that measure a person's regular information. Hereditary biometrics has, for the most part, used to validate and distinguish people by examining their physical attributes, for example, unique finger impression, eye iris, vein and so forth. Rather than utilizing a bank card, a camera presented at the ATM's would get pictures of countenances of clients, and separate them and the photographs of record holders in the database of banks to confirm the client's character. The motivation for driving this paper is to exhibit a Windows-based advancing application framework utilizing face certification checks.

We present a new method which generalizes subspace learning based on eigenvalue and generalized eigenvalue problems. This method, Roweis Discriminant Analysis (RDA) named after Sam Roweis, is a family of infinite number of algorithms where Principal Component Analysis (PCA), Supervised PCA (SPCA), and Fisher Discriminant Analysis (FDA) are special cases. One of the extreme special cases, named Double Supervised Discriminant Analysis (DSDA), uses the labels twice and it is novel. We propose a dual for RDA for some special cases. We also propose kernel RDA, generalizing kernel PCA, kernel SPCA, and kernel FDA, using both dual RDA and representation theory. Our theoretical analysis explains previously known facts such as why SPCA can use regression but FDA cannot, why PCA and SPCA have duals but FDA does not, why kernel PCA and kernel SPCA use kernel trick but kernel FDA does not, and why PCA is the best linear method for reconstruction. Roweisfaces and kernel Roweisfaces are also proposed generalizing eigenfaces, Fisherfaces, supervised eigenfaces, and their kernel variants. We also report experiments showing the effectiveness of RDA and kernel RDA on some benchmark datasets.

This work proposes an edge computing-based visitor identification system for elderly or dementia residents to recognize the visitors to their smart home quickly. The proposed system uses audio and lighting stimulation to trigger an associative recall in the subjects for recognizing familiar persons. Frontal face haar cascade classifier is used in extracting the facial portion of the captured visitor image. Classification of the visitor is done by eigenfaces approach using the extracted facial image. Playing relevant audio file on the smart speakers located in the house triggers associative recall in the elderly or dementia persons based on the principles of music therapy. Also, relevant lighting effects by hue lights in the house assist the subjects in recognizing the visitors quickly. The system realizes the benefits of music and lighting therapy in everyday living by identifying the familiar people and exploits the advantages of edge/fog computing.

Computer vision becomes a great area of research due to the huge availability of images and videos. For enhancement of security, biomedical imaging, or automation of identification, one may need useful tools to recognize images. One main problem of image data set is high dimensional, and it is very expensive to work with huge dimensions. In this paper, our main aim is to show a better dimension reduction process of high dimensional image data sets from several existing techniques. To verify it we start with most useful singular value decomposition to reduce the dimensionality of data to incorporate principal components. On the other hand, we classify data in advance to work out Fisher’s discriminant. From many real-world examples, we set a very well-known paradigm of analysis using Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) or Fisher Discriminant Analysis (FDA) and Simple Projection (SP) to recognize people from their facial images. We consider that we have some images of known people that can be used to compare and recognize new images (of the same set of face images). Moreover, we show graphical and tabular representation for average performance of correct recognition as well as analyze the effectiveness of three different machine learning techniques.

During the festival seasons, the crowds in the shopping malls become eight to ten times more than regular times. To impress maximum number of customers, business owners offer discounts, arrange raffle draws, game competitions, and many other events. They decorate their showrooms with eye-catching banners and lighting effects. These excellent marketing strategies work and increase customers’ engagements. However, the curious minds of children are attracted by these events more than those of adults. As a result, missing children in the shopping malls have become a common incident during festival seasons. Whenever a child is missing, shopping mall managements announce it though central sound system and normally if the people see the missing child, they bring it back to mall management’s office and the officials hand it over to the parents. This method depends on the generosity of the people and does not guarantee the return of the child. Moreover, the children kidnappers can take advantage of the announcement of missing child. In this paper, we are proposing a method that depends on eigenfaces which is fast, efficient yet simple to implement and ensure the recovery of the missing children in shortest possible time.

Facial Expression Recognition (FER) has become one of the most popular areas of research in computer vision and biometrics authentication and it has achieved a lot of enthusiasm from researchers. The Vision based Facial Expression Recognition system intends to classify the facial expression of a given image. In this paper, the proposed system automatically classifies the facial expression. The system is composed of feature extraction and expression classification. In preprocessing, Hybrid filter (Median and Gabor) and Histogram Equalizations, is used to reduce noise and enhance images. Feature extraction is to extract feature vectors from face images using the Eigenfaces approach, based on Principal Component Analysis (PCA). To classify facial expression, extracted feature vectors are fed into a Multiclass Support Vector Machine (Multi-SVM) classifier. Experiments are performed on the standard dataset of the Japanese Female Facial Expression (JAFFE) and achieved 80% accuracy. The proposed system showed satisfying performance comparing with other methods and effects state-of-the-art performance on the JAFFE dataset.

The methods and techniques of detection of the human face and facial recognition have presented a great impulse in recent years, thanks to the advance in areas such as artificial vision and machine learning. Although deep neural network techniques are in vogue, traditional techniques allow you to create applications that do not consume many resources from computing devices. In this research, we present a facial recognition system that implements the Eigenfaces method, developed in C # of Microsoft Visual Studio and open-source video processing libraries such as OpenCV as EmguCV. The application is divided into two sections: the first called register where, through an integrated camera, images of the user’s face or other means such as video and stored images are captured, and the second section is known as recognition where the user is compared with all the records of the data set, indicating whether this is registered and the recognition percentage. The project was implemented with a universe of the size of twenty-five users, of which six are men (24%) and nineteen are women (76%), developing tests for five weeks.

Facial recognition technologies are implemented in many areas, including but not limited to, citizen surveillance, crime control, activity monitoring, and facial expression evaluation. However, processing biometric information is a resource-intensive task that often involves third-party servers, which can be accessed by adversaries with malicious intent. Biometric information delivered to untrusted third-party servers in an uncontrolled manner can be considered a significant privacy leak (i.e. uncontrolled information release) as biometrics can be correlated with sensitive data such as healthcare or financial records. In this paper, we propose a privacy-preserving technique for “controlled information release”, where we disguise an original face image and prevent leakage of the biometric features while identifying a person. We introduce a new privacy-preserving face recognition protocol named PEEP ( P rivacy using E ig E nface P erturbation) that utilizes local differential privacy. PEEP applies perturbation to Eigenfaces utilizing differential privacy and stores only the perturbed data in the third-party servers to run a standard Eigenface recognition algorithm. As a result, the trained model will not be vulnerable to privacy attacks such as membership inference and model memorization attacks. Our experiments show that PEEP exhibits a classification accuracy of around 70% - 90% under standard privacy settings.

Emotion is equivalent to mood or state of human emotion that correlates with non-verbal behavior. Related literature shows that humans tend to give off a clue for a particular feeling through nonverbal cues such as facial expression. This study aims to analyze the emotion of students using Philippines-based corpus of a facial expression such as fear, disgust, surprised, sad, anger and neutral with 611 examples validated by psychology experts and results aggregates the final emotion, and it will be used to define the meaning of emotion and connect it with a teaching pedagogy to support decisions on teaching strategies. The experiments used feature extraction methods such as Haar-Cascade classifier for face detection; Gabor filter and eigenfaces API for features extraction; and support vector machine in training the model with 80.11% accuracy. The result was analyzed and correlated with the appropriate teaching pedagogies for educators and suggest that relevant interventions can be predicted based on emotions observed in a lecture setting or a class. Implementing the prototype in Java environment, it captured images in actual class to scale the actual performance rating and had an average accuracy of 60.83 %. It concludes that through aggregating the facial expressions of students in the class, an adaptive learning strategy can be developed and implemented in the classroom environment.

The present study shows the quantitative analysis of the evaluation of traditional methods of facial recognition versus the evaluation of methods based on deep learning, the contribution of this study lies in the conditions of the dataset to be used, which unlike the regular paradigms, it contains a limited size of images to train what the researcher has called a hostile training environment, which will serve as a reference to determine what technique and/or algorithm can work to recognize faces in which you do not have a lot of information to be processed and with a poor quality of the face image to be recognized, factors that in recent years are recurring in the search requests of missing persons or in the absence of the quality of certain security cameras.

This paper discusses the results of a study that aimed to develop an eigenface technique known as (PC) 2A that collect the image of the original face with its vertical and horizontal projections. The basic components of the image were analyzed in the image enrichment section. An evaluation of the proposed method demonstrates that it costs less than the standard eigenface technique. Moreover, the experimental results show that a front-end database that has a gray level for each person has one training image; thus, in terms of accuracy, it was possible to get a 3-5% result for the proposed (PC)2A, which is higher than the precision of the standard eigenface technique. The main objective of this paper is to demonstrate the weaknesses and strengthens of the facial recognition approach as an identifier known as eigenfaces. This aim was achieved by using the principal components analysis algorithm based on the images of previously stored training data. The outcomes show the strength of the proposed technique, in which it was possible to obtain accuracy results of up to 96%, which in turn provides support for developing the technique proposed in this paper in the future because this work is of great importance in the field of biological treatments, the need for which has significantly increased over the last 5 years.