• 한국정밀공학회지
• /
• 제32권4호
• /
• pp.377-386
• /
• 2015

Joint force/torque estimation by inverse dynamics is a traditional tool in biomechanical studies. Conventionally for this, kinematic data of human body is obtained by motion capture cameras, of which the bulkiness and occlusion problem make it hard to capture a broad range of movement. As an alternative, inertial motion sensing using cheap and small inertial sensors has been studied recently. In this research, the performance of inertial motion sensing especially to calculate inverse dynamics is studied. Kinematic data from inertial motion sensors is used to calculate ground reaction force (GRF), which is compared to the force plate readings (ground truth) and additionally to the estimation result from optical method. The GRF estimation result showed high correlation and low normalized RMSE(R=0.93, normalized RMSE<0.02 of body weight), which performed even better than conventional optical method. This result guarantees enough accuracy of inertial motion sensing to be used in inverse dynamics analysis.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.814-822
• /
• 2022

The deployment of sensors for Structural Health Monitoring requires a complicated network arrangement, ground truthing, and calibration for validating sensor performance periodically. Any conventional sensor on a structural element is also subjected to static and dynamic vertical loadings in conjunction with other environmental factors, such as brightness, noise, temperature, and humidity. A structural model with strain gauges was built and tested to get realistic sensory information. This paper investigates different deep learning architectures and algorithms, including unsupervised, autoencoder, and supervised methods, to benchmark blind drift calibration methods using deep learning. It involves a fully connected neural network (FCNN), a long short-term memory (LSTM), and a gated recurrent unit (GRU) to address the blind drift calibration problem (i.e., performing calibrations of installed sensors when ground truth is not available). The results show that the supervised methods perform much better than unsupervised methods, such as an autoencoder, when ground truths are available. Furthermore, taking advantage of time-series information, the GRU model generates the most precise predictions to remove the drift overall.

• 로봇학회논문지
• /
• 제18권2호
• /
• pp.182-188
• /
• 2023

Due to the acceleration of an aging society, the need for lower limb exoskeletons to assist gait is increasing. And for use in daily life, it is essential to have technology that can accurately estimate gait phase even in the walking environment and walking speed of the wearer that changes frequently. In this paper, we implement an LSTM-based gait phase estimation learning model by collecting gait data according to changes in gait speed in outdoor level ground and stair environments. In addition, the results of the gait phase estimation error for each walking environment were compared after learning for both max hip extension (MHE) and max hip flexion (MHF), which are ground truth criteria in gait phase divided in previous studies. As a result, the average error rate of all walking environments using MHF reference data and MHE reference data was 2.97% and 4.36%, respectively, and the result of using MHF reference data was 1.39% lower than the result of using MHE reference data.

• 천문학회보
• /
• 제40권1호
• /
• pp.61.3-62
• /
• 2015

Remote-sensing observation is one of the observation methods that provide valuable information, such as composition and surface physical conditions of solar system objects. The Hayabusa spacecraft succeeded in the first sample returning from a near-Earth asteroid, (25143) Itokawa. It has established a ground truth technique to connect between ordinary chondrite meteorites and S-type asteroids. One of the scientific observation instruments that Hayabusa carried, Asteroid Multi-band Imaging Camera(AMICA) has seven optical-near infrared filters (ul, b, v, w, x, p, and zs), taking more than 1400 images of Itokawa during the rendezvous phase. The reflectance of planetary body can provide valuable information of the surface properties, such as the optical aspect of asteroid surface at near zero phase angle (i.e. Sun-asteroid-observer's angle is nearly zero), light scattering on the surface, and surface roughness. However, only little information of the phase angle dependences of the reflectance of the asteroid is known so far. In this study, we investigated the phase angle dependences of Itokawa's surface to understand the surface properties in the solar phase angle of $0^{\circ}-40^{\circ}$ using AMICA images. About 700 images at the Hayabusa rendezvous phase were used for this study. In addition, we compared our result with those of several photometry models, Minnaert model, Lommel-Seeliger model, and Hapke model. At this conference, we focus on the AMICA's v-band data to compare with previous ground-based observation researches.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2002년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.734-738
• /
• 2002

We present spectral characteristics of tidal flat sediments and algal mat that were tested in the Gomso and Saemangum tidal flats, Korea. The objective of this study is to investigate the spectral reflectance and the radar scattering modeling in the tidal flats. Ground truth data obtained in the tidal flats include grain size, soil moisture content and its variation with time, surface roughness, chlorophyll, ground leveling, and field spectral reflectance measurement. The concept of an effective exposed area (EEA) is introduced to accommodate the effect of remnant surface water, and it seriously affects the reflection of short wavelength infrared and microwave. The nin size of 0.0625 mm has been normally used as a critical size of mud and sand discrimination. But we propose here that 0.25 mm is more practical grain size criterion to discriminate by remote sensing. Algal mat is the primary product in tidal flats, and it is found to be very important to understand spectral characteristics for tidal flat remote sensing. We have also conducted radar scattering modeling, and showed L-band HV-polarization would be the most effective combination.

• Water Engineering Research
• /
• 제1권3호
• /
• pp.209-222
• /
• 2000

A spectral formalism was applied to quantify the sampling errors due to spatial and/or temporal gaps in soil moisture measurements. The lack of temporal measurements of the two-dimensional soil moisture field makes it difficult to compute the spectra directly from observed records. Therefore, the space-time soil moisture spectra derived by stochastic models of rainfall and soil moisture was used in their record. Parameters for both models were tuned with Southern Great Plains Hydrology Experiment(SGP'97) data and the Oklahoma Mesonet data. The structure of soil moisture data is discrete in space and time. A design filter was developed to compute the sampling errors for discrete measurements in space and time. This filter has the advantage in its general form applicable for all kinds of sampling designs. Sampling errors of the soil moisture estimation during the SGP'97 Hydrology Experiment period were estimated. The sampling errors for various sampling designs such as satedlite over pass and point measurement ground probe were estimated under the climate condition between June and August 1997 and soil properties of the SGP'97 experimental area. The ground truth design was evaluated to 25km and 50km spatial gap and the temporal gap from zero to 5 days.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2003년도 Proceedings of ACRS 2003 ISRS
• /
• pp.37-37
• /
• 2003

Rice is one of the most important crop in the world and is a major export of Thailand. Optical sensors are not useful for rice monitoring, because most cultivated areas are often obscured by cloud during the growing period, especially in South East Asia. Spaceborne Synthetic Aperture Radar (SAR) such as RADARSAT, can see through regardless of weather condition which make it possible to monitor rice growth and to retrieve rice acreage, using the unique temporal signature of rice fields. This paper presents the result of a study of examining the backscatter behavior of rice using multi-temporal RADARSAT dataset. Ground measurements of paddy parameters and water and soil condition were collected. The ground truth information was also used to identify mature rice crops, orchard, road, residence, and aquaculture ponds. Land use class distributions from the RADARSAT image were analyzed. Comparison of the mean DB of each land use class indicated significant differences. Schematic representation of temporal backscatter of rice crop were plotted. Based on the study carried out in Pathum Thani Province test site, the results showed variation of sigma naught from first tillering vegatative phase until ripenning phase. It is suggested that at least, three radar data acquisitions taken at 3 stages of rice growth circle namely; those are at the beginning of rice growth when the field is still covered with water, in the ear differentiation period, and at the beginning of the harvest season, are required for rice monitoring. This pilot project was an experimental one aiming at future operational rice monitoring and potential yield predicttion.

• 인터넷정보학회논문지
• /
• 제21권6호
• /
• pp.23-31
• /
• 2020

본 연구는 철도표면상에 발생하는 노후 현상 중 하나인 결함 검출을 위해 학습데이터를 생성함으로써 결함 검출 모델에서 더 높은 점수를 얻기 위해 진행되었다. 철도표면에서 결함은 선로결속장치 및 선로와 차량의 마찰 등 다양한 원인에 의해 발생하고 선로 파손 등의 사고를 유발할 수 있기 때문에 결함에 대한 철도 유지관리가 필요 하다. 그래서 철도 유지관리의 자동화 및 비용절감을 위해 철도 표면 영상에 영상처리 또는 기계학습을 활용한 결함 검출 및 검사에 대한 다양한 연구가 진행되고 있다. 일반적으로 영상 처리 분석기법 및 기계학습 기술의 성능은 데이터의 수량과 품질에 의존한다. 그렇기 때문에 일부 연구는 일반적이고 다양한 철도표면영상의 데이터베이스를 확보하기위해 등간격으로 선로표면을 촬영하는 장치 또는 탑재된 차량이 필요로 하였다. 본연구는 이러한 기계적인 영상획득 장치의 운용비용을 감소시키고 보완하기 위해 대표적인 영상생성관련 딥러닝 모델인 생성적 적대적 네트워크의 기본 구성에서 여러 관련연구에서 제시된 방법을 응용, 결함이 있는 철도 표면 재생성모델을 구성하여, 전용 데이터베이스가 구축되지 않은 철도 표면 영상에 대해서도 결함 검출을 진행할 수 있도록 하였다. 구성한 모델은 상이한 철도 표면 텍스처들을 반영한 철도 표면 생성을 학습하고 여러 임의의 결함의 위치에 대한 Ground-Truth들을 만족하는 다양한 결함을 재 생성하도록 설계하였다. 재생성된 철도 표면의 영상들을 결함 검출 딥러닝 모델에 학습데이터로 사용한다. 재생성모델의 유효성을 검증하기 위해 철도표면데이터를 3가지의 하위집합으로 군집화 하여 하나의 집합세트를 원본 영상으로 정의하고, 다른 두개의 나머지 하위집합들의 몇가지의 선로표면영상을 텍스처 영상으로 사용하여 새로운 철도 표면 영상을 생성한다. 그리고 결함 검출 모델에서 학습데이터로 생성된 새로운 철도 표면 영상을 사용하였을 때와, 생성된 철도 표면 영상이 없는 원본 영상을 사용하였을 때를 나누어 검증한다. 앞서 분류했던 하위집합들 중에서 원본영상으로 사용된 집합세트를 제외한 두 개의 하위집합들은 각각의 환경에서 학습된 결함 검출 모델에서 검증하여 출력인 픽셀단위 분류지도 영상을 얻는다. 이 픽셀단위 분류지도영상들과 실제 결함의 위치에 대한 원본결함 지도(Ground-Truth)들의 IoU(Intersection over Union) 및 F1-score로 평가하여 성능을 계산하였다. 결과적으로 두개의 하위집합의 텍스처 영상을 이용한 재생성된 학습데이터를 학습한 결함 검출모델의 점수는 원본 영상만을 학습하였을 때의 점수보다 약 IoU 및 F1-score가 10~15% 증가하였다. 이는 전용 학습 데이터가 구축되지 않은 철도표면 영상에 대해서도 기존 데이터를 이용하여 결함 검출이 상당히 가능함을 증명하는 것이다.

• Agricultural and Biosystems Engineering
• /
• 제4권2호
• /
• pp.50-57
• /
• 2003

This study was conducted to develop software which provides automatic site-specific field data acquisition, data processing, data mapping and management for precision agriculture. The developed software supports acquisition and processing of both digital and analog data streams. The architecture was object-oriented and each component in the architecture was developed as a separate class. In precision agriculture research, the laborious task of manual ground-truth data collection will be avoided using the developed software. In addition, gathering high-density data eliminates the need for interpolation of values for un-sampled areas. This software shows good potential for expansion and compatibility for variable-rate-application (VRA). The FIM (Field Information Management) computer program provides the user with an easy-to-follow process for field information management for precision agriculture.

• International Journal of Contents
• /
• 제14권1호
• /
• pp.1-6
• /
• 2018

Text, as one of the most influential inventions of humanity, has played an important role in human life since ancient times. The rich and precise information embodied in text is very useful in a wide range of vision-based applications such as the text data extracted from images that can provide information for automatic annotation, indexing, language translation, and the assistance systems for impaired persons. Therefore, natural-scene text detection with active research topics regarding computer vision and document analysis is very important. Previous methods have poor performances due to numerous false-positive and true-negative regions. In this paper, a fully-convolutional-network (FCN)-based method that uses supervised architecture is used to localize textual regions. The model was trained directly using images wherein pixel values were used as inputs and binary ground truth was used as label. The method was evaluated using ICDAR-2013 dataset and proved to be comparable to other feature-based methods. It could expedite research on text detection using deep-learning based approach in the future.