• Journal of Information Processing Systems
• /
• v.16 no.1
• /
• pp.6-29
• /
• 2020

Biometrics identification using multiple modalities has attracted the attention of many researchers as it produces more robust and trustworthy results than single modality biometrics. In this paper, we present a novel multimodal recognition system that trains a deep learning network to automatically learn features after extracting multiple biometric modalities from a single data source, i.e., facial video clips. Utilizing different modalities, i.e., left ear, left profile face, frontal face, right profile face, and right ear, present in the facial video clips, we train supervised denoising auto-encoders to automatically extract robust and non-redundant features. The automatically learned features are then used to train modality specific sparse classifiers to perform the multimodal recognition. Moreover, the proposed technique has proven robust when some of the above modalities were missing during the testing. The proposed system has three main components that are responsible for detection, which consists of modality specific detectors to automatically detect images of different modalities present in facial video clips; feature selection, which uses supervised denoising sparse auto-encoders network to capture discriminative representations that are robust to the illumination and pose variations; and classification, which consists of a set of modality specific sparse representation classifiers for unimodal recognition, followed by score level fusion of the recognition results of the available modalities. Experiments conducted on the constrained facial video dataset (WVU) and the unconstrained facial video dataset (HONDA/UCSD), resulted in a 99.17% and 97.14% Rank-1 recognition rates, respectively. The multimodal recognition accuracy demonstrates the superiority and robustness of the proposed approach irrespective of the illumination, non-planar movement, and pose variations present in the video clips even in the situation of missing modalities.

• Journal of Microbiology and Biotechnology
• /
• v.22 no.12
• /
• pp.1644-1652
• /
• 2012

Iron plays a key role in host-pathogen interactions. Microbial pathogens require iron for survival and virulence, whereas mammalian hosts sequester and withhold iron as a means of nutritional immunity. We previously identified two paralogous genes, CFT1 and CFT2, which encode homologs of a fungal iron permease, Cft1 and Cft2, respectively, in the human fungal pathogen Cryptococcus neoformans. Cft1 was shown to play a role in the high-affinity reductive iron uptake system, and was required for transferrin utilization and full virulence in mammalian hosts. However, no role of Cft2 has been suggested yet. Here, we identified the third gene, CFT3, that produces an additional fungal iron permease homolog in C. neoformans, and we also generated the cft3 mutant for functional characterization. We aimed to reveal distinct functions of Cft1, Cft2 and Cft3 by analyzing phenotypes of the mutants lacking CFT1, CFT2 and CFT3, respectively. The endogenous promoter of CFT1, CFT2 and CFT3 was replaced with the inducible GAL7 promoter in the wild-type strain or in the cft1 mutant for gain-of-function analysis. Using these strains, we were able to find that CFT2 is required for growth in low-iron conditions in the absence of CFT1 and that overexpression of CFT2 compensates for deficiency of the cft1 mutant in iron uptake and various cellular stress conditions. However, unlike CFT2, no clear phenotypic characteristic of the cft3 mutant and the strain overexpressing CFT3 was observed. Overall, our data suggested a redundant role of Cft2 in the high-affinity iron uptake and stress responses in C. neoformans.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.35 no.4
• /
• pp.347-352
• /
• 2007

This paper describes the design and implementation of PFM(Proto Flight Model, PFM) of DRE(Data Receiving Equipment, DRE) for Science and Technology Satellite 2(STSAT-2) and the results of integration performance test. DRE components are X-band receiver, DCE(Data Combine Equipment, DCE) and RAC(Receiving and Archiving Computer, RAC). DCE consists of I&Q data combiner and ECL signal distributor. RAC consists of DRC(Data Receiving Card) and ST2RAS(STSAT-2 Receiving and Archinving Software). X-band receiver receives 10Mbps QPSK I, Q satellite data and sends the data to DCE. DRC stores the I&Q combine data from DCE to RAID. The pre-processing program sorts and stores to satellite status data and payload data. The performance of DRE in the functional and space environments test satisfies the requirements of STSAT-2.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.11
• /
• pp.1086-1090
• /
• 2010

In recent, as many applications of sensor networks increase, various techniques have been studied to efficiently operate network systems. The query optimization scheme that is one of such techniques has been studied to reduce the data transmission cost. The data transmission is of great importance to the energy consumption of sensor networks. In this paper, we propose an energy-efficient multiple queries processing scheme by sharing sensor readings for multiple queries, when they are occurred in sensor networks. The proposed scheme reduces unnecessary data transmissions among the sensor nodes by intuitively identifying their locations using the grid structure. It also efficiently shares the data by recognizing the redundant regions of sensor nodes. In order to show the superiority of the proposed scheme, we compare it with the existing scheme in various experiments. As the result, the proposed scheme reduces about 65% energy consumption over the existing scheme.

• The KIPS Transactions:PartB
• /
• v.12B no.2 s.98
• /
• pp.131-136
• /
• 2005

Motion Estimation (ME) has been developed to remove redundant data contained in a sequence of image. And ME is an important part of video encoding systems, since it can significantly affect the qualify of an encoded sequences. Generally, ME consists of two stages, the integer pixel motion estimation and the half pixel motion estimation. Many methods have been developed to reduce the computational complexity at the integer pixel motion estimation. However, the studies are needed at the half pixel motion estimation to reduce the complexity. In this paper, a method based on the correlations between integer pixel motion vectors and half pixel motion vectors is proposed for the half pixel motion estimation. The proposed method has less computational complexity than the full half pixel search method (FHSM) that needs the bilinear interpolation of half pixels and examines nine half pixel points to the find the half pixel motion vector. Experimental results show that the speedup improvement of the proposed method over FHSM can be up to $2.5\~80$ times faster and the image quality degradation is about to $0.07\~0.69(dB)$.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.19 no.5
• /
• pp.80-86
• /
• 2005

Induction motors are a critical component of many industrial processes and are frequently integrated in commercially available equipment. Safety, reliability, efficiency, and performance are some of the major concerns of induction motor applications. Fault tolerant control (FTC) strives to make the system stable and retain acceptable performance under the system faults. All present FTC method can be classified into two groups. The first group is based on fault detection and diagnostics (FDD). The second group is includes of FDD and includes methods such as integrity control, reliable stabilization and simultaneous stabilization. This paper presents the fundamental FDD-based FTC methods, which are capable of on-line detection and diagnose of the induction motors. Therefore, our group has developed the embedded distributed fault tolerant and fault diagnosis system for industrial motor. This paper presents its architecture. These mechanisms are based on two 32-bit DSPs and each TMS320F2407 DSP module processes the stator current, voltage, temperatures, vibration signal of the motor.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.5
• /
• pp.1049-1058
• /
• 2017

The rapid evolution of mobile technologies and proliferation of mobile devices have created a new channel for marketing by mobile advertising. As mobile advertising is a close relative to online advertising, it also has similar problems such as advertisement injections (Ad injections). Users are exposed to unwanted advertisements and redundant web traffic by injected ads can cause additional charges of mobile devices. Although mobile ad injection can cause many problems it has been merely studied. In this paper, we analyze ad injection activities by mobile applications that exploit a legitimate application (Naver mobile application). We reverse-engineered 2 mobile applications and find out characteristics of mobile ad injections. We compare mobile ad injections with online ad injections and suggest feasible mitigations.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.8 no.2
• /
• pp.245-249
• /
• 2004

In this paper, we have proposed a clock recovery algorithm of Orthogonal Frequency Division Multiplexing/Quadrature Phase Shift Keying Modulation-Digital Microwave Radio(OFDM/QPSK-DMR) system using Band Limited-Pulse Shaping Filter(BL-PSF) and compared the clock phase error variance of OFDM/QPSK-DMR system with that of single carrier DMR system. The OFDM/QPSK-DMR system using windowing method requires training sequence or Cyclic Prefix (CP) to synchronize the clock phase of received signal. But transmit efficient is increased in our proposed DMR system because of no using redundant data such as training sequence or CP. The proposed clock recovery algorithm is simply realized in the OFDM/QPSK-DMR system using BL-PSF. The simulation results confirm that the proposed clock recovery algorithm has the same clock phase error variance performance in a single carrier DR system under Additive White Gaussian Noise(AWGN) environment.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.10
• /
• pp.6056-6065
• /
• 2014

The aim of this study was to develop a scale of the Korea citizen's intentionality out of their inborn ability and mission. To set up the concept of the above intentionality, there have been many academic and religious studies for reference in addition to the existing productivity, Korea elderly successful aging scale, and the scale of life qualities for review. The first 57 items for questionnaires were created thereafter. The final 27 items were confirmed, which were constantly modified by a professor of statistics and corresponding author in effective response and in removing the redundant items after the first items had been reduced to 44 preliminary items through the many an item-by-item discussions with pastors, theologians and social workers. A survey was conducted around the metropolitan area and 380 records were collected. Factor analysis to verify the validity of the measurement tool was completed, and showed that there were 6 sub-factors; "recognition of being", "autonomous life", "self-completion", "initiative", "acceptance", and "economic feasibility". In addition, reliability analysis was implemented and found to be reliable because the Cronbach's ${\alpha}$ value was 0.6 or more.

• Journal of KIISE:Databases
• /
• v.33 no.2
• /
• pp.151-162
• /
• 2006

RDF Semantics, one of W3C Recommendations, provides the RDFS entailment rules, which are used for the RDFS inference. Sesame, which is well known RDF repository, supports the RDBMS-based RDFS inference using the forward-chaining strategy. Since inferencing in the forward-chaining strategy is performed in the data loading time, the data loading time in Sesame is slow down be inferencing. In this paper, we propose the order scheme for applying the RDFS entailment rules to improve inference performance. The proposed application order makes the inference process terminate without repetition of the process for most cases and guarantees the completeness of inference result. Also the application order helps to reduce redundant results during the inference by predicting the results which were made already by previously applied rules. In this paper, we show that our approaches can improve the inference performance with comparisons to the original Sesame using several real-life RDF datasets.