• Journal of Drive and Control
• /
• v.17 no.3
• /
• pp.1-8
• /
• 2020

Recently, a special purpose machine with two manipulators and quadruped crawler system has been developed for rapid life-saving and initial restoration work at disaster sites. This special purpose machine provides the driver with various environmental recognition functions for accurate and rapid task determination. In particular, the human detection technology assists the driver in poor working conditions such as low-light, dust, water vapor, fog, rain, etc. to prevent secondary human accidents when moving and working. In this study, a human detection module is developed to be mounted on a special purpose machine. A thermal sensor and CCD camera were used to detect victims and nearby workers in response to the difficult environmental conditions present at disaster sites. The performance of various AI-based life detection algorithm were verified and then applied to the task of detecting various objects with different postures and exposure conditions. In addition, image visibility improvement technology was applied to further improve the accuracy of human detection.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1117-1121
• /
• 2020

In this work, it was confirmed that a SnO2 catalyst deposited by an atomic layer deposition(ALD) process can be employed in AlGaN/GaN heterostructure FET to detect NO2 gas. The fabricated HFET sensors on AlGaN/GaN-on-Si platform demonstrated that the devices with or without n-situ SiN have sensitivity of 5.5 % and 38 % at 200 ℃, respectively with response to 100 ppm-NO2.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.429-431
• /
• 2022

하지 웨어러블 로봇의 근력 보조 성능을 극대화하기 위해서는 착용자의 보행 상태를 인식하는 보행 위상 추정 기술이 필수적으로 요구된다. 본 논문에서는 착용자의 보행 속도 변화 및 착용자 간 보행 특성 차이에도 강인하게 보행 위상을 추정할 수 있는 LSTM 기반 보행 위상 강건 인식 기술을 개발하였다. 웨어러블 고관절 보조 로봇을 착용한 총 5명의 트레드밀 및 실외 overground의 보행 센서 정보를 바탕으로 학습을 수행하였다. 저속 및 고속 보행을 포함한 다양한 보행 속도에서 정밀한 보행 위상 추정이 가능한 웨어러블 센서 조합을 도출하였고, 보행 위상 인식 정밀성은 5-Fold Cross Validation 기준 RMSE 약 1.68% 수준의 결과를 얻을 수 있었다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.1
• /
• pp.179-186
• /
• 2015

This paper proposes the target path detection algorithm using statistical characteristics of an activated time lag along a moving path of target from a neighboring sensor in PDR(Pulse Doppler Radar) sensor node environment based on USN(Ubiquitous Sensor Network) with a limitation detecting only an existence of moving target. In the proposed algorithm, detection and non-detection time lag obtained from the experimental data are used. The experimental data are through repetitive action of each 500 times about three path scenarios such as passing in between two sensors, moving parallel to two sensors, and turning through two sensors. From this experiments, error detection percentages of three path scenarios are 5.67%, 5.83%, and 7.17%, respectively. They show that the proposed algorithm can exactly detect a target path using the limited PDR sensor nodes.

• Journal of Advanced Navigation Technology
• /
• v.23 no.1
• /
• pp.69-76
• /
• 2019

In this paper, a design method for an enhanced-sensitivity microwave sensor based on microstrip defected ground structure was studied for the permittivity measurement of planar dielectric substrates. The proposed sensor was designed by modifying the ridge structure of an H-shaped aperture into the shape of a capacitor symbol. The sensitivity of the proposed sensor was compared with that of a conventional sensor based on a double-ring complementary split ring resonator(DR-CSRR). Two sensors were designed and fabricated on a 0.76-mm-thick RF-35 substrate so that the transmission coefficient would resonate at 1.5 GHz in the absence of the substrate under test. Five types of taconic substrates with a relative permittivity ranging from 2.17 to 10.2 were selected asthe substrate under test. Experiment results show that the sensitivity of the proposed sensor, which is measured by the shift in the resonant frequency of the transmission coefficient, is 1.31 to 1.62 times higher than that of the conventional DR-CSRR-based sensor.

• Journal of Advanced Navigation Technology
• /
• v.23 no.5
• /
• pp.415-423
• /
• 2019

In this paper, a design method for a high-sensitivity microstrip patch sensor antenna (MPSA) loaded with a bent-slot was studied for the permittivity measurement. The bent-slot similar to a single-ring complementary split ring resonator was added along a radiating edge of the patch in order to enhance the sensitivity to the permittivity. The sensitivity of the proposed MPSA was compared with that of a conventional rectangular MPSA and a thin rectangular-slotted MPSA. Three MPSAs were designed and fabricated on a 0.76-mm-thick RF-35 substrate so that the input reflection coefficient would resonate at 2.5 GHz in the absence of the superstrate under test. When five different Taconic substrates with a relative permittivity ranging from 2.17 to 10.2 were used as the superstrate under test, experiment results show that the sensitivity of the proposed MPSA, which is measured by the shift in the resonant frequency of the input reflection coefficient, is 4.1 to 6.1 times higher than that of the conventional MPSA.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.2
• /
• pp.379-384
• /
• 2022

In a wireless sensor network, a large number of sensor nodes are deployed in an environment where direct access is difficult. It is difficult to supply power, such as replacing the battery or recharging it. It is very important to use the energy with the sensor node. Therefore, an important consideration to increase the lifetime of the network is to minimize the energy consumption of each sensor node. If the energy of the wireless sensor node is exhausted and discharged, it cannot function as a sensor node. Therefore, it is a method proposed in various protocols to minimize the energy consumption of nodes and maintain the network for a long time. We consider the center point and residual energy of the cluster, and the plot point and K-means (WSN suggests optimal clustering). We want to evaluate the performance of the KOCED protocol. We compare protocols to which the K-means algorithm, one of the latest machine learning methods, is applied, and present performance evaluation factors.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_1
• /
• pp.511-521
• /
• 2022

Cloud detection and removal from satellite images is an essential process for topographic observation and analysis. Threshold-based cloud detection techniques show stable performance because they detect using the physical characteristics of clouds, but they have the disadvantage of requiring all channels' images and long computational time. Cloud detection techniques using deep learning, which have been studied recently, show short computational time and excellent performance even using only four or less channel (RGB, NIR) images. In this paper, we confirm the performance dependence of the deep learning network according to the heterogeneous learning dataset with different resolutions. The DeepLabV3+ network was improved so that channel features of cloud detection were extracted and learned with two published heterogeneous datasets and mixed data respectively. As a result of the experiment, clouds' Jaccard index was low in a network that learned with different kind of images from test images. However, clouds' Jaccard index was high in a network learned with mixed data that added some of the same kind of test data. Clouds are not structured in a shape, so reflecting channel features in learning is more effective in cloud detection than spatial features. It is necessary to learn channel features of each satellite sensors for cloud detection. Therefore, cloud detection of heterogeneous sensors with different resolutions is very dependent on the learning dataset.

• Journal of Sensor Science and Technology
• /
• v.11 no.3
• /
• pp.155-162
• /
• 2002

The sensing properties of carbon monooxide were investigated as a function of mixing ratio and the lamination structure of 3mol% ZnO-doped $SnO_2$ and 3mol% $SnO_2$-doped ZnO. The lamination structures were fabricared monolayer, double layer, and hetero layer of $SnO_2$, Zno, and theirs mixture composition using thick film process. There was no second phase by the reaction of $SnO_2$ and ZnO. The conductance was decreased by the addition of ZnO in $SnO_2$, but it was increased with the addition of $SnO_2$ in ZnO. The conductance was increased with temperature and the inlet of CO. There was no improvement of sensitivity in the structure of mono- and double-layer. The hetero-layer structure, however, of $SnO_2$ 3ZnO-ZnO $3SnO_2$ showed the higher resistivity and the highest sensitivity. Ohmic characteristics was confirmed by the linear properties for I-V measurements.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.172-174
• /
• 2019

본 논문에서는 사용자가 자신의 움직임 및 심장 박동 상태를 모니터링 하기 위한 생체 인터넷 (Internet of Biometry: IoB) 기반의 HW/SW (hardware/software) 플랫폼 (platform)을 제안한다. 제안하는 시스템은 모션 센서 (motion sensor) 또는 심박 (heart rate) 센서와 같이 사용자의 생체 정보를 수집할 수 있는 센서를 사용한다. 또한, 마이크로프로세서 (microprocessor)를 사용하여 센서로부터 수집된 데이터를 사용자에게 필요한 생체 정보로 변환하고, 블루투스 (Bluetooth) 통신을 이용하여 사용자의 스마트폰 앱 (smartphone application)으로 변환한 생체 정보를 전달한다. 스마트폰 앱은 수신한 생체 정보를 디스플레이 (display)함으로써, 사용자가 자신의 상태를 모니터링 (monitoring) 할 수 있다. 제안한 시스템을 사용하여 해양 레포츠 (leisure sports) 등과 같은 활동을 하는 사람들이 자신의 몸 상태를 스스로 확인할 수 있고, 사고 예방의 효과를 얻을 수 있다.