Papers by Keyword: Face Recognition

Abstract: In this article, the training parameters of cascade classifiers for specific targets (such as human faces and palms) are obtained by using the program training function provided by OpenCV and adjusting parameters such as the number of positive and negative samples and ratios to achieve better recognition results. Finally, the aiming and locking system for specific targets is constructed by combining recognition, algorithm, and robot control capabilities.

Abstract: Nowadays face recognition system is widely used in every field of computer vision applications such as Face lock-in smartphones, surveillance, smart attendance system, and driverless car technology. Because of this, the demand for face recognition systems is increasing day by day in the research field. The aim of this project is to develop a system that will recommend music based on facial expressions. The face-recognition system consists of object detection and identifying facial features from input images, and the face recognition system can be made more accurate with the use of convolutional neural networks. Layers of convolution neural network are used for the expression detection and are optimized with Adam to reduce overall loss and improve accuracy. YouTube song playlist recommendation is an application of a face recognition system based on a neural network. We use streamlit-webrtc to design the web frame for the song recommendation system. For face detection, we used the Kaggle-FER2013 dataset, and images in the dataset are classified into seven natural emotions of a person. The system captures the emotional state of a person in real-time and generates a playlist of youtube songs based on that emotion.

Abstract: Adaboost Viola-Jones method is indeed a profound discovery in detecting face images mainly because it is fast, light and one of the easiest methods of detecting face images among other techniques of face detection. Viola Jones uses Haar wavelet filter to detect face images and it produces almost 80%accuracy of face detection. This paper discusses proposed methodology and algorithms that involved larger library of filters used to create more discrimination features among the images by processing the proposed 15 Haar rectangular features (an extension from 4 Haar wavelet filters of Viola Jones) and used them in multiple adaptive ensemble process of detecting face image. After facial detection, the process continues with normalization processes by applying feature extraction such as PCA combined with LDA or LPP to extract our week learners’ wavelet for more classification features. Upon the process of feature extraction proposed feature selection to index these extracted data. These extracted vectors are used for training and creating MADBoost (Multiple Adaptive Diversified Boost)(an improvement of Adaboost, which uses multiple feature extraction methods combined with multiple classifiers) is able to capture, recognize and distinguish face image (s) faster. MADBoost applies the ensemble approach with better weights for classification to produce better face recognition results. Three experiments have been conducted to investigate the performance of the proposed MADBoost with three other classifiers, Neural Network (NN), Support Vector Machines (SVM) and Adaboost classifiers using Principal Component Analysis (PCA) as the feature extraction method. These experiments were tested against obstacles of POIES (Pose, Obstruction, Illumination, Expression, Sizes). Based on the results obtained, Madboost is found to be able to improve the recognition performance in matching failures, incorrect matching, matching success percentages and acceptable time taken to perform the classification task.

Abstract: Face detection is the primary task in building a vision-based human-computer interaction system and in special applications such as face recognition, face tracking, face identification, expression recognition and also content-based image retrieval. A potent face detection system must be able to detect faces irrespective of illuminations, shadows, cluttered backgrounds, orientation and facial expressions. In previous literature, many approaches for face detection had been proposed. However, face detection in outdoor images with uncontrolled illumination and images with complex background are still a serious problem. Hence, in this paper, we had proposed a Geometric Skin Colour (GSC) method for detecting faces accurately in real world image, under capturing conditions of both indoor and outdoor, and with a variety of illuminations and also in cluttered backgrounds. The selected method was evaluated on two different face video smartphone databases and the obtained results proved the outperformance of the proposed method under the unconstrained environment of these databases.

Abstract: In this paper, an embedded facial recognition system whose platform consists of pcDuono-V2 board with ARM-processor inside and a Linux-kernel-based operating system, Ubuntu, is implemented. A camera is set up on the platform to take human face images. A facial recognition program consisting of AdaBoost algorithm, Haar-like features, integral image method, and cascade classifiers is utilized to recognize images. The AdaBoost algorithm is a modified Boosting algorithm, which is a machine learning algorithm for training cascade stronger classifiers based on Haar-like features, where Haar-like features are the foundation of the recognition. An integral image method is used to speed up the calculation of corresponding rectangle feature values for Haar-like features. The whole facial recognition comprises facial training procedures and recognition procedures. In facial training procedures, sufficient amounts of positive and negative picture samples are necessary for getting Haar-like features to the recognition system. AdaBoost algorithm is then used to the system for training cascade stronger classifiers which are the detection tools in recognition procedures. While in facial recognition procedures, after getting the Haar-like features for the target images or pictures, cascade stronger classifiers work to detect and recognize. According to the experimental results, the resultant embedded system can recognize the experimental subjects in one second for every our considered situations, which assures the real-time performance.

Abstract: The aim of the project is to implement a facial recognition system for access control to enter a room. The facial image captured by a webcam and then be detected/tracked using Haar face tracking algorithm. Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) algorithm have been used for face recognition. The system was tested with 10 users from the member of a laboratory room. Each user registered 100 images for training of the PCA and LDA. The recognition rate achieved using PCA was 70% and 97% for LDA.

Abstract: In pattern recognition such as face recognition, the recognition result is not only limited by the quality and quantity of samples, but also limited by the extracted principal components. For improving the quality and quantity of training samples and for extracting more efficient principal components, this paper presents a recognition method combing the increased virtual samples and kernel principal component analysis (KPCA), which doubly weakens the influence of nonlinear factors on face recognition. New database is generated with the pose-changed and the mirror-like virtual images. Then KPCA is used for dimension reduction and feature extraction. The shortest Euclidean distance is applied to measure similarity. A series of experiments are conducted in the ORL and YALE face database and the experimental results show the efficiency of the proposed method.

Abstract: In the field of object recognition, the SIFT feature is known to be a very successful local invariant descriptor and has wide application in different domains. However it also has some limitations, for example, in the case of facial illumination variation or under large tilt angle, the identification rate of the SIFT algorithm drops quickly. In order to reduce the probability of mismatching pairs, and improve the matching efficiency of SIFT algorithm, this paper proposes a novel feature matching algorithm. The basic idea is taking the successful-matched SIFT feature points as the training samples to establish a space mapping model based on BP neural network. Then, with the help of this model, the estimated coordinate of the corresponding SIFT feature point in the candidate image is predicted. Finally search the possible matching points around the coordinate. The experiment results show that using the prediction model, the number of mismatching points can be reduced effectively and the number of correct matching pairs increases at the same time

Abstract: This paper proposed a new face recognition algorithm based on Haar-Like features and Gentle Adaboost feature selection via sparse representation. Firstly, All the images including face images and non face images are normalized to size and then Haar-Like features are extracted . The number of Haar-Like features can be as large as 12,519. In order to reduce the feature dimension and retain the most effective features for face recognition, Gentle Adaboost algorithm is used for feature selection. Selected features are used for face recognition via sparse representation classification (SRC) algorithm. Testing experiments were carried out on the AR database to test the performance of the new proposed algorithm. Compared with traditional algorithms like NS, NN, SRC, and SVM, the new algorithm achieved a better recognition rate. The effect of face recognition rate changing with feature dimension showed that the new proposed algorithm performed a higher recognition rate than SRC algorithm all the time with the increasing of feature dimension, which fully proved the effectiveness and superiority of the new proposed algorithm.

Abstract: In order to achieve fast and accurate face recognition, firstly, this paper extracts face contour lines based on face contour feature to achieve treat a human face shape classification, thereby reducing the measured face search range. Secondly, using the method of Gabor wavelet to extract local texture on face when range was approximately determined, then matching with the face library whose range has been narrowed. Experiments show that this parallel approach not only achieved rapid recognition, but also achieve a high recognition rate due to a combination of the two methods.