Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725643
Quoc-Dung Nguyen, Tin Huynh, Tu-Anh Nguyen-Hoang
Job recommender systems are designed to suggest a ranked list of jobs that could be associated with employee's interest. Most of existing systems use only one approach to make recommendation for all employees, while a specific method normally is good enough for a group of employees. Therefore, this study proposes an adaptive solution to make job recommendation for different groups of user. The proposed methods are based on employee clustering. Firstly, we group employees into different clusters. Then, we select a suitable method for each user cluster based on empirical evaluation. The proposed methods include CB-Plus, CF-jFilter and HyR-jFilter have applied for different three clusters. Empirical results show that our proposed methods is outperformed than traditional methods.
{"title":"Adaptive methods for job recommendation based on user clustering","authors":"Quoc-Dung Nguyen, Tin Huynh, Tu-Anh Nguyen-Hoang","doi":"10.1109/NICS.2016.7725643","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725643","url":null,"abstract":"Job recommender systems are designed to suggest a ranked list of jobs that could be associated with employee's interest. Most of existing systems use only one approach to make recommendation for all employees, while a specific method normally is good enough for a group of employees. Therefore, this study proposes an adaptive solution to make job recommendation for different groups of user. The proposed methods are based on employee clustering. Firstly, we group employees into different clusters. Then, we select a suitable method for each user cluster based on empirical evaluation. The proposed methods include CB-Plus, CF-jFilter and HyR-jFilter have applied for different three clusters. Empirical results show that our proposed methods is outperformed than traditional methods.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134516007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725641
Thu A. Pham, H. Pham, Hai-Chau Le, N. Dang
In this paper, we investigate the performance of a millimeter-wave (MMW) radio-over-fiber (RoF) link, which is used to provide high data-rate and flexible connections in next-generation broadband optical-wireless access networks. The mixed optical fiber/wireless channel is assumed to be a dispersive and fading channel. Other physical layer impairments including noise and nonlinear distortion are also included in our analysis. The numerical results show that fiber dispersion and fading are key factors that limit the link performance. Therefore, dispersion shifted fibers should be used to reduce the effect of dispersion so as to improve the bit-error rate. To mitigate the effect of multipath fading, the remote radio heads (RRHs) or users should be located as near the remote antenna unit (RAU) as possible.
{"title":"Performance analysis of MMW/RoF link in broadband optical-wireless access networks","authors":"Thu A. Pham, H. Pham, Hai-Chau Le, N. Dang","doi":"10.1109/NICS.2016.7725641","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725641","url":null,"abstract":"In this paper, we investigate the performance of a millimeter-wave (MMW) radio-over-fiber (RoF) link, which is used to provide high data-rate and flexible connections in next-generation broadband optical-wireless access networks. The mixed optical fiber/wireless channel is assumed to be a dispersive and fading channel. Other physical layer impairments including noise and nonlinear distortion are also included in our analysis. The numerical results show that fiber dispersion and fading are key factors that limit the link performance. Therefore, dispersion shifted fibers should be used to reduce the effect of dispersion so as to improve the bit-error rate. To mitigate the effect of multipath fading, the remote radio heads (RRHs) or users should be located as near the remote antenna unit (RAU) as possible.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121615667","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725637
C. T. Nguyen, Dang Tuan Nguyen
In this paper, an argument-based method of answering why-questions in Vietnamese is presented. This method is developed in different way from many approaches which use cue phrases of causal relation to find the answers for why-questions. In this method, the arguments is extracted firstly, then the causal part and consequential part of every argument are split in order to index the consequential part. When a why-question is asked, the asking information is extracted and used to search for the reason, then the reason is used to identify the paragraph which can be used to answer the question. For evaluation, an experiment with keyword-based information retrieval and simple argument collecting process is conducted to show the applicability of the method.
{"title":"Towards an argument-based method for answering why-question in Vietnamese language","authors":"C. T. Nguyen, Dang Tuan Nguyen","doi":"10.1109/NICS.2016.7725637","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725637","url":null,"abstract":"In this paper, an argument-based method of answering why-questions in Vietnamese is presented. This method is developed in different way from many approaches which use cue phrases of causal relation to find the answers for why-questions. In this method, the arguments is extracted firstly, then the causal part and consequential part of every argument are split in order to index the consequential part. When a why-question is asked, the asking information is extracted and used to search for the reason, then the reason is used to identify the paragraph which can be used to answer the question. For evaluation, an experiment with keyword-based information retrieval and simple argument collecting process is conducted to show the applicability of the method.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129888967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725666
Thanh Tung Vu, H. H. Kha
In this paper, precoding techniques are investigated for secure energy-efficient multiple-input multiple-output (MIMO) systems in the presence of an eavesdropper. To tackle the non-convexity of the secure energy efficiency maximization (SEEM) problem, we employ branch-and-reduce-and-bound (BRB) approach to obtain the globally optimal solution. It is observed that the BRB algorithm suffers from highly computational cost, its globally optimal solution is importantly served as a benchmark for the performance evaluation of the suboptimal algorithms. This motivates us to develop a low-complexity approach using the well-known zero-forcing (ZF) technique to cancel the wiretapped signal. In ZF based method, the SEEM problem can be transformed to a concave-convex fractional one and, then, be solved by the combination of the Dinkelbach and bisection search algorithm. Simulation results reveal that the ZF based method can converge fast and obtain a sub-optimal performance in compared with the optimal solution of the BRB method. The ZF based scheme also outperforms the conventional secrecy rate maximization precoder design in terms of the energy efficiency.
{"title":"Optimal precoder designs for energy-efficiency maximization in secure MIMO systems","authors":"Thanh Tung Vu, H. H. Kha","doi":"10.1109/NICS.2016.7725666","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725666","url":null,"abstract":"In this paper, precoding techniques are investigated for secure energy-efficient multiple-input multiple-output (MIMO) systems in the presence of an eavesdropper. To tackle the non-convexity of the secure energy efficiency maximization (SEEM) problem, we employ branch-and-reduce-and-bound (BRB) approach to obtain the globally optimal solution. It is observed that the BRB algorithm suffers from highly computational cost, its globally optimal solution is importantly served as a benchmark for the performance evaluation of the suboptimal algorithms. This motivates us to develop a low-complexity approach using the well-known zero-forcing (ZF) technique to cancel the wiretapped signal. In ZF based method, the SEEM problem can be transformed to a concave-convex fractional one and, then, be solved by the combination of the Dinkelbach and bisection search algorithm. Simulation results reveal that the ZF based method can converge fast and obtain a sub-optimal performance in compared with the optimal solution of the BRB method. The ZF based scheme also outperforms the conventional secrecy rate maximization precoder design in terms of the energy efficiency.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127210277","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725662
Tan-Sy Nguyen, Thai-Hoang Huynh
The paper presents a new control architecture consisting of two control loops to balance an Inertia Wheel Pendulum (IWP) of which the wheel is driven by a DC motor. The outer control loop uses a LQR controller to calculate the required torque to balance the IWP around the upright position based on the measurement of the pendulum angle and angular velocity. The inner control loop uses a PI controller to calculate the voltage applied to the DC motor such that the torque provided by the motor meets the requirement of the outer loop. Matlab simulation and real hardware experiments have been carried out to validate the effectiveness of the proposed method.
{"title":"Study on a two-loop control architecture to balance an inertia wheel pendulum","authors":"Tan-Sy Nguyen, Thai-Hoang Huynh","doi":"10.1109/NICS.2016.7725662","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725662","url":null,"abstract":"The paper presents a new control architecture consisting of two control loops to balance an Inertia Wheel Pendulum (IWP) of which the wheel is driven by a DC motor. The outer control loop uses a LQR controller to calculate the required torque to balance the IWP around the upright position based on the measurement of the pendulum angle and angular velocity. The inner control loop uses a PI controller to calculate the voltage applied to the DC motor such that the torque provided by the motor meets the requirement of the outer loop. Matlab simulation and real hardware experiments have been carried out to validate the effectiveness of the proposed method.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132192687","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725635
Mompoloki Pule, A. Yahya, J. Chuma
Wireless Sensor Networks (WSNs) provide reliable and affordable infrastructure for various applications ranging from environmental monitoring, industrial automation, telemedicine, to military surveillance. These networks allow monitoring processes to be conducted remotely, in real-time and with minimal human intervention. Water quality monitoring is essential to the provision of clean and safe water. In Botswana, Water Utilities Corporation (WUC), has adopted a conventional like process that employs several independent hand held meters for sampling different water quality parameters manually. However, this process has proved to be ineffective since it is expensive, time consuming, requires considerable human resource capacity and lacks real-time results to promote proactive response to water contamination. In this work, an integrated WSN based solution that employs Libelium's Waspmote platform has been developed. The sensor node is based on ATMEL's ATmega 1281 8-bit low power microcontroller which can operate for more than a year without recharging its battery. Results have shown that the proposed system managed to sample associated water quality parameters in real time with the same accuracy as currently adopted methods. In addition, the system is autonomous, easy to install, highly scalable and has a relatively high sampling rate which improves the temporal resolution of the monitoring process.
{"title":"A wireless sensor network solution for monitoring water quality in Botswana","authors":"Mompoloki Pule, A. Yahya, J. Chuma","doi":"10.1109/NICS.2016.7725635","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725635","url":null,"abstract":"Wireless Sensor Networks (WSNs) provide reliable and affordable infrastructure for various applications ranging from environmental monitoring, industrial automation, telemedicine, to military surveillance. These networks allow monitoring processes to be conducted remotely, in real-time and with minimal human intervention. Water quality monitoring is essential to the provision of clean and safe water. In Botswana, Water Utilities Corporation (WUC), has adopted a conventional like process that employs several independent hand held meters for sampling different water quality parameters manually. However, this process has proved to be ineffective since it is expensive, time consuming, requires considerable human resource capacity and lacks real-time results to promote proactive response to water contamination. In this work, an integrated WSN based solution that employs Libelium's Waspmote platform has been developed. The sensor node is based on ATMEL's ATmega 1281 8-bit low power microcontroller which can operate for more than a year without recharging its battery. Results have shown that the proposed system managed to sample associated water quality parameters in real time with the same accuracy as currently adopted methods. In addition, the system is autonomous, easy to install, highly scalable and has a relatively high sampling rate which improves the temporal resolution of the monitoring process.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134464263","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725675
Van-Tinh Nguyen, V. Dao, T. Phan
This paper designs and implements a codec system using the Cyclic code on FPGA. The encoding system was based on the principle of dividing circuits and the decoding system was based on the principle of the Meggitt decoder. This work proposes the look-up table (LUT) method for the decoding system. The implementation results from FPGA show that the proposed decoding method has exactly resulted. In addition, the proposed cyclic decoder core using the look-up table method has lower resource and number of cycles compared to the cyclic decoder core using the Meggitt method.
{"title":"Hardware implementation of cyclic codes error correction on FPGA","authors":"Van-Tinh Nguyen, V. Dao, T. Phan","doi":"10.1109/NICS.2016.7725675","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725675","url":null,"abstract":"This paper designs and implements a codec system using the Cyclic code on FPGA. The encoding system was based on the principle of dividing circuits and the decoding system was based on the principle of the Meggitt decoder. This work proposes the look-up table (LUT) method for the decoding system. The implementation results from FPGA show that the proposed decoding method has exactly resulted. In addition, the proposed cyclic decoder core using the look-up table method has lower resource and number of cycles compared to the cyclic decoder core using the Meggitt method.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121203397","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725648
C. Pham-Quoc, Biet Nguyen-Hoang, T. N. Thinh
This paper proposes a reconfigurable heterogeneous multicore architecture to integrate multiple DDoS defense mechanisms for DDoS protection. The architecture allows multiple cooperating DDoS mitigation techniques to classify incoming network packets. The proposed architecture consists of two separated partitions: static and dynamic. The static partition includes packet pre-processing and post-processing modules while the DDoS filtering techniques are implemented on the dynamic partition. These filtering techniques can be implemented by either hardware custom computing cores or general purpose soft processors or both. In all cases, these DDoS filtering computing cores can be updated or changed at runtime or design time. We implement our first prototype system with Hop-count filtering and Ingress/Engress filtering techniques using Xilinx Virtex 5xc5vtx240t FPGA device. The synthesis results show that the system can work at up to 116.782MHz while utilizing about 41% LUTs, 47% Registers, and 53% Block Memory of the available hardware resources. The system achieves the detection rate of 100% with the false negative rate at 0% and false positive rate closed to 0.74%. The prototype system achieves packet decoding throughput at 9.869 Gbps in half-duplex mode and 19.738 Gbps in full-duplex mode.
{"title":"A reconfigurable heterogeneous multicore architecture for DDoS protection","authors":"C. Pham-Quoc, Biet Nguyen-Hoang, T. N. Thinh","doi":"10.1109/NICS.2016.7725648","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725648","url":null,"abstract":"This paper proposes a reconfigurable heterogeneous multicore architecture to integrate multiple DDoS defense mechanisms for DDoS protection. The architecture allows multiple cooperating DDoS mitigation techniques to classify incoming network packets. The proposed architecture consists of two separated partitions: static and dynamic. The static partition includes packet pre-processing and post-processing modules while the DDoS filtering techniques are implemented on the dynamic partition. These filtering techniques can be implemented by either hardware custom computing cores or general purpose soft processors or both. In all cases, these DDoS filtering computing cores can be updated or changed at runtime or design time. We implement our first prototype system with Hop-count filtering and Ingress/Engress filtering techniques using Xilinx Virtex 5xc5vtx240t FPGA device. The synthesis results show that the system can work at up to 116.782MHz while utilizing about 41% LUTs, 47% Registers, and 53% Block Memory of the available hardware resources. The system achieves the detection rate of 100% with the false negative rate at 0% and false positive rate closed to 0.74%. The prototype system achieves packet decoding throughput at 9.869 Gbps in half-duplex mode and 19.738 Gbps in full-duplex mode.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116820757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725633
Dang The Hung, Tran Trung Duy, D. Trinh, Vo Nguyen Quoc Bao, Tan Hanhv
This paper investigates impact of hardware impairments on the probability of Non-zero Secrecy Capacity (NSC) of multi-hop relay networks. In the considered protocol, multiple eavesdroppers attempt to overhear the data that is relayed from a source to a destination via multiple intermediate relay nodes. To enhance the secrecy performance, the multi-hop randomize-and-forward (RF) strategy is used to confuse the eavesdroppers. We derive exact expressions of the NSC probability in integral forms over Rayleigh fading channels. Moreover, these formulas are then expressed by sums of infinite series of exponential functions and exponential integral functions. Finally, Monte Carlo simulations are performed to verify our derivations.
{"title":"Impact of hardware impairments on secrecy performance of multi-hop relay networks in presence of multiple eavesdroppers","authors":"Dang The Hung, Tran Trung Duy, D. Trinh, Vo Nguyen Quoc Bao, Tan Hanhv","doi":"10.1109/NICS.2016.7725633","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725633","url":null,"abstract":"This paper investigates impact of hardware impairments on the probability of Non-zero Secrecy Capacity (NSC) of multi-hop relay networks. In the considered protocol, multiple eavesdroppers attempt to overhear the data that is relayed from a source to a destination via multiple intermediate relay nodes. To enhance the secrecy performance, the multi-hop randomize-and-forward (RF) strategy is used to confuse the eavesdroppers. We derive exact expressions of the NSC probability in integral forms over Rayleigh fading channels. Moreover, these formulas are then expressed by sums of infinite series of exponential functions and exponential integral functions. Finally, Monte Carlo simulations are performed to verify our derivations.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131923621","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2016-09-01DOI: 10.1109/NICS.2016.7725652
Huu-Tuan Nguyen, T. Huong
One of the most critical tasks in building a gender classification is how to describe the human face as a highly discriminative feature vector. To this end, in this paper we introduce a new handcrafted feature extraction method for unconstrained gender classification problem. From one input face image, we generate its smaller version and apply two LPQ operators on both of them. We then combine the obtained LPQ features with the SIFT feature extracted from the given image to constitute a global facial description. In the classification stage, the binary SVM classifier is used for determining the gender of the test images. To evaluate the recognition performance of the proposed methods, we carry out experiments upon two widely used unconstrained face databases Adience and LFW. The results show that our approach attains good classification rates (96.51% and 80.5% on LFW and Adience databases, respectively) and can be comparable with state-of-the-art systems.
{"title":"Unconstrained gender classification by multi-resolution LPQ and SIFT","authors":"Huu-Tuan Nguyen, T. Huong","doi":"10.1109/NICS.2016.7725652","DOIUrl":"https://doi.org/10.1109/NICS.2016.7725652","url":null,"abstract":"One of the most critical tasks in building a gender classification is how to describe the human face as a highly discriminative feature vector. To this end, in this paper we introduce a new handcrafted feature extraction method for unconstrained gender classification problem. From one input face image, we generate its smaller version and apply two LPQ operators on both of them. We then combine the obtained LPQ features with the SIFT feature extracted from the given image to constitute a global facial description. In the classification stage, the binary SVM classifier is used for determining the gender of the test images. To evaluate the recognition performance of the proposed methods, we carry out experiments upon two widely used unconstrained face databases Adience and LFW. The results show that our approach attains good classification rates (96.51% and 80.5% on LFW and Adience databases, respectively) and can be comparable with state-of-the-art systems.","PeriodicalId":347057,"journal":{"name":"2016 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science (NICS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134353780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：