• 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.

• The KIPS Transactions:PartC
• /
• v.9C no.1
• /
• pp.1-8
• /
• 2002

Digital image watermarking is a technique for the purpose of protecting the ownership of the image by embedding proprietary watermarks in a digital image. It is required for the digital image watermarking scheme to pursue the robustness against water marking attacks and the perceptual Invisibility more than usual in steganography area, to guarantee not a hidden watermarking algorithm but the publicity of water-marking algorithm details and hidden use of key, which can protect the unauthorized user access from detection. In this paper we propose a new copyright watermarking scheme, which is barred on one-way hash functions using ElGamal functions and modular operations. ElGamal functions are widely used in cryptographic systems. Our watermarking scheme is robust against LSB(least significant bit) attacks and gamma correction attack, and also perceptually invisible. We demonstrate the characteristics of our proposed watermarking scheme through experiments. It is necessary to proceed as the future work the algorithm of achieving at the same time both the pseudo-randomness for the steno-key generation and the asymmetric-key generation.

• Journal of Sensor Science and Technology
• /
• v.2 no.1
• /
• pp.19-27
• /
• 1993

Capacitive pressure sensor for low pressure measurements has been fabricated by using $n^{+}$ epitaxial layer electrochemical etching stop and glass-to-silicon electrostatic bonding technique. The sensor had hybrid configuration of a sensor chip, which consists of sensor capacitor and reference capacitor, and two output signal detection IC chips. A fabricated sensor, with a $1.0{\times}1.0 mm^{2}$ square size and a $10{\mu}m$ thick flat diaphragm, showed a 7.1 pF zero pressure capacitance, and 5.2 % F.S, sensitivity in 10 KPa pressure range. By using a capacitance to voltage converter, the thermal zero shift of 0.051 %F.S./$^{\circ}C$ and the thermal sensitivity shift of 0.12 %F.S./$^{\circ}C$ for temperature range of $5{\sim}45^{\circ}C$ were obtained.

• Environmental Engineering Research
• /
• v.17 no.4
• /
• pp.205-210
• /
• 2012

Hydroperoxyl radical/superoxide anion radical ($HO_2{\cdot}/O_2^-{\cdot}$, $pK_a$=4.8) as an intermediate is of considerable importance in oxidation processes. Hence, the method of detecting $HO_2{\cdot}/O_2^-{\cdot}$ with high sensitivity is necessary to be developed. To achieve this objective, this study newly employed terephthalate (TA) as a probe for the measurement of $HO_2{\cdot}/O_2^-{\cdot}$ in the kinetic method presented in our previous study. This method was based on the hydroxylation of TA to produce mainly hydroxyterephthalic acid or hydroxyterephthalate (OHTA), which was analyzed by fluorescence detection (${\lambda}_{ex}$=315nm, ${\lambda}_{ex}$=425nm). The life-time of $HO_2{\cdot}/O_2^-{\cdot}$ and its concentration formed from the photolysis technique of $H_2O_2$ were reported in this study. At range of pH 2-10, the life-time of $HO_2{\cdot}/O_2^-{\cdot}$ was 51-422 sec. In particular, an increase in the life-time with pH was observed. The sensitivities of the kinetic method by using TA were always higher with 1.7-2.5 times at pH 8.0 than those by using benzoic acid. From these results, this study can contribute to understanding the basic functions of $HO_2{\cdot}/O_2^-{\cdot}$ in oxidation processes.

• Analytical Science and Technology
• /
• v.14 no.6
• /
• pp.499-503
• /
• 2001

A new simple, rapid and sensitive gas chromatographic technique for the determination of bisphenol-A in can materials, which is the major material of epoxy resin and polycarbonate polymer, is proposed. This method is characterised by derivatization of the bisphenol-A with a acylating reagent forming the acetate derivative to optimize the chromatographic property. The detection of bisphenol-A is performed based on GC/MS (gas chromatography/mass spectrometry). Several beverages were analyzed by the proposed method for the determination of bisphenol-A Bisphenol-A was assayed the range of $0.11{\sim}11.40{\mu}g/can$ from the can materials.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.44 no.1
• /
• pp.12-22
• /
• 2016

Lidar-based hazard avoidance landing system for lunar lander calculates hazard cost with respect to the desired local landing area in order to identify hazard and designate safe landing point where the cost is minimum basically using slope and roughness of the landing area. In this case, if the parameters are only considered, chosen landing target can be designated near hazard threatening the lander. In order to solve this problem and select optimal safe landing point, hazard cost based on relative distance to hazard should not be considered as well as cost based on terrain parameters. In this paper, the effect of hazard cost based on relative distance to hazard on safe landing performance was analyzed and it was confirmed that landing site designation with two relative distances to hazard results in the best safe landing performance by an experiment using three-dimensional depth camera.

• Journal of Broadcast Engineering
• /
• v.13 no.2
• /
• pp.236-244
• /
• 2008

We propose a new building detection and description algorithm for Lidar data and photogrammetric imagery using color segmentation, line segments matching, perceptual grouping. Our algorithm consists of two steps. In the first step, from the initial building regions extracted from Lidar data and the color segmentation results from the photogrammetric imagery, we extract coarse building boundaries based on the Lidar results with split and merge technique from aerial imagery. In the secondstep, we extract precise building boundaries based on coarse building boundaries and edges from aerial imagery using line segments matching and perceptual grouping. The contribution of this algorithm is that color information in photogrammetric imagery is used to complement collapsed building boundaries obtained by Lidar. Moreover, linearity of the edges and construction of closed roof form are used to reflect the characteristic of man-made object. Experimental results on multisensor data demonstrate that the proposed algorithm produces more accurate and reliable results than Lidar sensor.

• Journal of Biomedical Engineering Research
• /
• v.24 no.4
• /
• pp.339-346
• /
• 2003

MRI provides anatomical structure information with superb spatial resolution that can be utilized in clinical surgeries. Advanced image processing techniques in conjunction with the MRI-guided surgery is expected to be of great importance in brain surgeries in the near future. In this paper, we introduce an image-guided surgery technique using the stereo matching method. To perform image-guided biopsy operations, we made MRI markers, camera markers and a detection probe marker. To evaluate the accuracy of the image-guided system. we made a silicone phantom. Using the phantom and markers, we have performed MRI-guided experiments with a 1.5 Tesla MRI system. It has been verified from phantom experiments that our system has a positioning error less than 1.5%. Compared with other image guided surgery system, our system shows better positioning accuracy.

• Proceedings of the Korean Magnetic Resonance Society Conference
• /
• 2002.08a
• /
• pp.70-71
• /
• 2002

Exchange-coupled cluster of transition-metal ions are relevant to many different scientific areas, ranging from chemistry to solid-state physics, biology, material science and has been the subject of much research in recent years(1,2). Single crystal EPR spectroscopy works as a very effective tool for the measurement of J values for small exchange interactions. This makes EPR technique very suitable for detection of weak exchange coupling transmitted over long distances via extended atomic and melecular bridges. Large polyoxometallates (3) may provide ideal structural environments for the study of interactions between paramagnetic ions. The detailed nature of magnetic interaction (positive sign and magnitude of J~0.006 $cm^{-1}$ /) was clearly determined for di-copper(II) system by single crystal EPR spectroscopy (4). The single-triplet (S-T) transitions are forbidden by different symmetries of the wave functions. However, when the singlet ground state is mixed into triplet states, the S-T transitions can be allowed and observed as weak lines. These weak S-T lines are positioned symmetrically with respect to the main transitions in the distance equals to 2J from the center of the spectrum. This lines allow one to determine the J-value with very high accuracy when │J│ ＜ hv 0.32 $cm^{-1}$ /. Unfortunately, the S-T transitions in the single crystal were detected by EPR method only in a few complexes until now. We have measured single-triplet transition lines for several magnetically coupled cluster systems and determined their J values accurately. The temperature dependency of J was studied by monitoring the changes in S-T.

• Journal of Navigation and Port Research
• /
• v.39 no.6
• /
• pp.479-484
• /
• 2015

Network RTK is a precise positioning technique using carrier phase correction data from reference stations within the network, and is constantly being researched for improved performance. However, the study for the system accuracy has been performed but system integrity research has not been done as much as system accuracy, because network RTK has been mainly used on surveying for static or kinematic positioning. In this paper, adequate monitoring system for network RTK is designed as basis research for integrity monitoring on network RTK. To this, fault tree on network RTK is analyzed, and a countermeasure is prepared to detect and identify the each fault items. Based these algorithms, monitoring system to use on central processing facility is designed for network RTK service.