• Journal of Software Assessment and Valuation
• /
• v.16 no.2
• /
• pp.63-68
• /
• 2020

This paper proposes a face authentication system based on deep learning framework. The proposed system is consisted of face region detection and feature extraction using deep learning algorithm, and performed the face authentication using joint-bayesian matrix learning algorithm. The performance of proposed paper is evaluated by various face database , and the face image of one person consists of 2 images. The face authentication algorithm was performed by measuring similarity by applying 2048 dimension characteristic and combined Bayesian algorithm through Deep Neural network and calculating the same error rate that failed face certification. The result of proposed paper shows that the proposed system using deep learning and joint bayesian algorithms showed the equal error rate of 1.2%, and have a good performance compared to previous approach.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.46 no.1
• /
• pp.55-67
• /
• 2023

Lightweight face recognition models, as one of the most popular and long-standing topics in the field of computer vision, has achieved vigorous development and has been widely used in many real-world applications due to fewer number of parameters, lower floating-point operations, and smaller model size. However, few surveys reviewed lightweight models and reimplemented these lightweight models by using the same calculating resource and training dataset. In this survey article, we present a comprehensive review about the recent research advances on the end-to-end efficient lightweight face recognition models and reimplement several of the most popular models. To start with, we introduce the overview of face recognition with lightweight models. Then, based on the construction of models, we categorize the lightweight models into: (1) artificially designing lightweight FR models, (2) pruned models to face recognition, (3) efficient automatic neural network architecture design based on neural architecture searching, (4) Knowledge distillation and (5) low-rank decomposition. As an example, we also introduce the SqueezeFaceNet and EfficientFaceNet by pruning SqueezeNet and EfficientNet. Additionally, we reimplement and present a detailed performance comparison of different lightweight models on the nine different test benchmarks. At last, the challenges and future works are provided. There are three main contributions in our survey: firstly, the categorized lightweight models can be conveniently identified so that we can explore new lightweight models for face recognition; secondly, the comprehensive performance comparisons are carried out so that ones can choose models when a state-of-the-art end-to-end face recognition system is deployed on mobile devices; thirdly, the challenges and future trends are stated to inspire our future works.

• Tribology and Lubricants
• /
• v.36 no.2
• /
• pp.75-81
• /
• 2020

In this paper, we present an experimental investigation of the leakage and endurance performances of mechanical face seals in a 75-tonf turbopump for the Korea Space Launch Vehicle II first-stage engine. A mechanical face seal is used between the fuel pump and turbine to prevent mixing of the fuel and turbine gas. However, excessive leakage occurs through the carbon attached to the mechanical face seal bellows. To reduce this leakage, we redesign the mechanical face seal such that the contact area between the fuel and carbon is reduced, height of the carbon nose is reduced, and stiffness of the bellows is increased. Then, we conduct static and dynamic leakage tests and endurance tests to compare the performances of the original and modified mechanical face seals. The investigation of the leakage of the old and new mechanical face seals confirms that the leakage performance is significantly improved, by 80%, in the new design in comparison with the old design. The endurance tests demonstrate that the average wear rate of carbon in the new mechanical face seal is 0.1094 ㎛/s. The service lifetime is predicted to be 4,200 s, which is 28 times greater than the requirement. Finally, we present a new mechanical face seal in a 75-tonf turbopump, and perform a validation test in the real-propellant test facility at the NARO Space Center. Based on the test results, we can confirm that the modified mechanical face seal works well under real operating conditions.

• The Journal of Korea Robotics Society
• /
• v.4 no.3
• /
• pp.192-200
• /
• 2009

In this paper, we report the progress in the development of performance evaluation method for detection algorithms of face region and facial components. This paper aims to provide a standardized evaluation method for general approach in face recognition application as a potential component in futuristic intelligent robot systems. From an image capture process to the retrieval of face-related information, all the necessary steps are shown with examples.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.17 no.3
• /
• pp.121-132
• /
• 2021

The purpose of this study is to explore ways to improve the quality of real-time video lectures in a non-face-to-face environment using IPA (Importance-Performance Analysis). Recently, due to the impact of COVID-19 in universities, all remote classes are being implemented, so research is needed to raise learner awareness. Accordingly, factor analysis, mean analysis, correspondence analysis, and IPA analysis were performed based on the data of 632 students who responded from March 21 to June 30, 2021 for learners of K University in Chungbuk. First, overall satisfaction was low compared to importance, and the difference in system perception was the largest. Second, the difference in learner perception of real-time video lectures through the IPA matrix showed that the system error and screen cutoff were the largest. Third, the difficulty of lecture content, task and test feedback, etc. are classified. Accordingly, the satisfaction of real-time video lectures in non-face-to-face environments is low, suggesting that school-level support for quality improvement to improve learner satisfaction in non-face-to-face environments and the role of instructors are needed to improve learners' academic achievement.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.1
• /
• pp.35-43
• /
• 2010

1D PCA utilized in the face appearance-based face recognition methods such as eigenface-based face recognition method may lead to less face representative power and more computational cost due to the resulting 1D face appearance data vector of high dimensionality. To resolve such problems of 1D PCA, 2D PCA-based face recognition methods had been developed. However, the face representation model obtained by direct application of 2D PCA to a face image set includes both face common features and face distinctive identity features. Face common features not only prevent face recognizability but also cause more computational cost. In this paper, we first develope a model of a face distinctive identity feature subspace separated from the effects of face common features in the face feature space obtained by application of 2D PCA analysis. Then, a novel robust face recognition based on the face distinctive identity feature subspace model is proposed. The proposed face recognition method based on the face distinctive identity feature subspace shows better performance than the conventional PCA-based methods (1D PCA-based one and 2D PCA-based one) with respect to recognition rate and processing time since it depends only on the face distinctive identity features. This is verified through various experiments using Yale A and IMM face database consisting of face images with various face poses under various illumination conditions.

• The Journal of the Korea Contents Association
• /
• v.9 no.2
• /
• pp.115-124
• /
• 2009

The popular eigenfaces-based face recognition among proposed face recognition methods utilizes the eigenfaces obtained from applying PCA to a training face image set. Thus, it may not achieve a reliable performance under illumination environments different from that of training face images. In this paper, we propose an illumination-separate eigenfaces-based face recognition method, which excludes the effects of illumination as much as possible. The proposed method utilizes the illumination-separate eigenfaces which is obtained by orthogonal decomposition of the eigenface space of face model image set with respect to the constructed face illumination subspace. Through experiments, it is shown that the proposed face recognition method based on the illumination-separate eigenfaces performs more robustly under various illumination environments than the conventional eigenfaces-based face recognition method.

• Journal of KIISE
• /
• v.41 no.9
• /
• pp.666-673
• /
• 2014

This paper proposes a scalable face recognition method for unconstrained face databases, and shows a simple experimental result. Existing face recognition research usually has focused on improving the recognition rate in a constrained environment where illumination, face alignment, facial expression, and background is controlled. Therefore, it cannot be applied in unconstrained face databases. The proposed system is face feature extraction algorithm for unconstrained face recognition. First of all, we extract the area that represent the important features(landmarks) in the face, like the eyes, nose, and mouth. Each landmark is represented by a high-dimensional LBP(Local Binary Pattern) histogram feature vector. The multi-scale LBP histogram vector corresponding to a single landmark, becomes a low-dimensional face feature vector through the feature reduction process, PCA(Principal Component Analysis) and LDA(Linear Discriminant Analysis). We use the Rank acquisition method and Precision at k(p@k) performance verification method for verifying the face recognition performance of the low-dimensional face feature by the proposed algorithm. To generate the experimental results of face recognition we used the FERET, LFW and PubFig83 database. The face recognition system using the proposed algorithm showed a better classification performance over the existing methods.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.477-480
• /
• 2018

Viola and Jones object detection algorithm is a representative face detection algorithm. The algorithm uses Haar-like features for face expression and uses a cascade-Adaboost algorithm consisting of strong classifiers, a linear combination of weak classifiers for classification. This algorithm requires several parameter settings for its implementation and the set values affect its performance. This paper analyzes face detection performance according to the parameters set in the algorithm.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.847-850
• /
• 2000

This paper proposes an image compressing and template matching algorithm for face image using GHA (Generalized Hebbian Algorithm). GHA is a part of PCA (Principal Component Analysis), that has single-layer perceptrons and operates and self-organizing performance. We used this algorithm for feature extraction of face shape, and our simulations verify the high performance for the proposed method. The shape for face in the fact that the eigenvector of face image can be efficiently represented as a coefficient that can be acquired by a set of basis is to compress data of image. From the simulation results, the mean PSNR performance is 24.08[dB] at 0.047bpp, and reconstruction experiment shows that good reconstruction capacity for an image that not joins at leaning.