• Korean Journal of Remote Sensing
• /
• v.36 no.4
• /
• pp.515-525
• /
• 2020

This study investigates the potential of bidirectional long short-term memory (Bi-LSTM) for efficient modeling of temporal information in crop classification using multitemporal remote sensing images. Unlike unidirectional LSTM models that consider only either forward or backward states, Bi-LSTM could account for temporal dependency of time-series images in both forward and backward directions. This property of Bi-LSTM can be effectively applied to crop classification when it is difficult to obtain full time-series images covering the entire growth cycle of crops. The classification performance of the Bi-LSTM is compared with that of two unidirectional LSTM architectures (forward and backward) with respect to different input image combinations via a case study of crop classification in Anbadegi, Korea. When full time-series images were used as inputs for classification, the Bi-LSTM outperformed the other unidirectional LSTM architectures; however, the difference in classification accuracy from unidirectional LSTM was not substantial. On the contrary, when using multitemporal images that did not include useful information for the discrimination of crops, the Bi-LSTM could compensate for the information deficiency by including temporal information from both forward and backward states, thereby achieving the best classification accuracy, compared with the unidirectional LSTM. These case study results indicate the efficiency of the Bi-LSTM for crop classification, particularly when limited input images are available.

• Journal of the Korea Safety Management & Science
• /
• v.17 no.4
• /
• pp.143-151
• /
• 2015

The purpose of this study was to investigate temporal and spatial variations, and moments of the lower extremities of gait while playing the game with smartphone under different curb-heights. Ten male elementary school students(from 10 years to 13 years old) participated in this study. Twelve infrared cameras(Oqus-500) and two force plates(9260AA) were used for collecting data and these were processed via Visual 3D software. In conclusion, with or without smartphone and with different curb-heights, the spatial and temporal parameters of walking were not the same and coefficients of variations were not consistent. The maximum joint moments of the lower extremities with or without smartphone were not statistically significant but those of hip and ankle joint were statistically significant with regard to the different heights of the curbs.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.6
• /
• pp.47-53
• /
• 2007

True motion estimation is necessary for deinterlacing, frame-rate conversion, and film judder compensation. There have been several block-based approaches to find true motion vectors by tracing minimum sum-of-absolute-difference (SAD) values by considering spatial and temporal consistency. However, the algorithms cannot find robust motion vectors when the texture of objects is changed. To find the robust motion vectors in the region, a recursive vector selection scheme and an adaptive weighting parameter are proposed. Previous frame vectors are recursively averaged to be utilized for motion error region. The weighting parameter controls fidelity to input vectors and the recursively averaged ones, where the input vectors come from the conventional estimators. If the input vectors are not reliable, then the mean vectors of the previous frame are used for temporal consistency. Experimental results show more robust motion vectors than those of the conventional methods in time-varying texture objects.

• Physical Therapy Korea
• /
• v.13 no.4
• /
• pp.79-86
• /
• 2006

The purpose of the current experiment was to describe interlimb coordination when swing limb conditions are being manipulated by constraining step length or by adding a 5 or 10 pound weight to the swing limb distally. Subjects were asked to begin walking with the right limb to land on the primary target (normal step length) that is 10 cm in diameter. However, if, during movement, the light was illuminated, then the subject had to step on one of the secondary targets (long and short step length). These three step length conditions were repeated while wearing a 5 pound ankle weight and then when wearing a 10 pound ankle weight. Ground reaction force (GRF) data indicated that there were changes in the forces and slopes of the swing and stance Fx GRFs. Long stepping subjects had to increase the propulsive force required to increase step length. Consequently, swing and stance toe-off greatly increased in the long step length condition. Short step length subjects had to adequately adjust step length, which decreased the speed of gait initiation. Loading the swing limb decreased the force and slope of the swing limb. Swing and stance toe-off was longest for the long step length condition, but there was a small difference of temporal events between no weight and weight condition. It appears that subjects modulated GRFs and temporal events differently to achieve the peak acceleration force of the swing and stance limb in response to different tasks. The findings from the current study provide preliminary data, which can be used to further investigate how we modulate forces during voluntary movement from a quiet stance. This information may be important if we are to use this or a similar task to evaluate gait patterns of the elderly and patient populations.

• Journal of Korea Water Resources Association
• /
• v.44 no.12
• /
• pp.941-954
• /
• 2011

In this study, past method and recent method for flood discharge with domestic multi-purpose dams in Korea were compared and analyzed with respect to the scale of watershed. Rainfall depth, temporal pattern, rainfall excess, rainfall-runoff model, parameter estimation and base flow were selected as the principal factors affecting flood discharge and effects on flood discharge were analyzed quantitatively by using sensitivity analysis. The results showed that the flood discharges calculated by past and recent method increased and decreased with a wide range of discharge with respect to the scale of watershed. The reason for decrease of flood discharge is the exchange of temporal pattern of rainfall and the principal reasons for increase of flood discharge are the increase of rainfall depth by unusual weather phenomena and the difference of estimation method for parameters of unit hydrograph.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.2
• /
• pp.401-406
• /
• 2008

Fall event detection is one of the most common problems for elderly people, especially those living alone because falls result in serious injuries such as joint dislocations, fractures, severe head injuries or even death. In order to prevent falls or fall-related injuries, several previous methods based on video sensor showed low fall detection rates in recent years. To improve this problem and outperform the system performance, this paper presented a novel approach for fall event detection in the elderly using a subtraction between successive difference images and temporal templates in real time environment. The proposed algorithm obtained the successful detection rate of 96.43% and the low false positive rate of 3.125% even though the low-quality video sequences are obtained by a USB PC camera sensor. The experimental results have shown very promising performance in terms of high detection rate and low false positive rate.

• Korean Journal of Remote Sensing
• /
• v.19 no.6
• /
• pp.465-478
• /
• 2003

This paper presents two methods for automatic estimation of threshold values in unsupervised change detection of multi-temporal remote sensing images. The proposed methods consist of two analytical steps. The first step is to compute the parameters of a 3-component Gaussian mixture model from difference or ratio images. The second step is to determine a threshold value using Bayesian rule for minimum error. The first method which is an extended version of Bruzzone and Prieto' method (2000) is to apply an Expectation-Maximization algorithm for estimation of the parameters of the Gaussian mixture model. The second method is based on an iterative thresholding algorithm that successively employs thresholding and estimation of the model parameters. The effectiveness and applicability of the methods proposed here were illustrated by two experiments and one case study including the synthetic data sets and KOMPSAT-1 EOC images. The experiments demonstrate that the proposed methods can effectively estimate the model parameters and the threshold value determined shows the minimum overall error.

• Journal of Convergence for Information Technology
• /
• v.8 no.4
• /
• pp.247-253
• /
• 2018

This study was designed to analyze the difference in attendance and academic achievement based on spatio-temporal data relate to e-Learning patterns of university e-Learning learners. This study collected e-Learning data from 68 e-Learning classes, 13,611 learners during 3 years. Collected data were analyzed by t-test and two-way ANOVA. Major study findings were as follows. Firstly, e-Learning learners in school received higher than those of learners outside school both in attendance and academic achievement, while that academic achievement showed statistical significance. Secondly, the attendance and academic achievement by the day was in the order of e-Learning learners mainly in the morning, those in the afternoon and those at night, in addition there was statistical significance. Lastly e-Learning learners in the weekdays appeared higher than those of learners in the weekends both in attendance and academic achievement, also both of them showed statistical significance.

• Journal of Korean Society on Water Environment
• /
• v.36 no.6
• /
• pp.508-520
• /
• 2020

Dam reservoirs have been reported to contribute significantly to global carbon emissions, but unlike natural lakes, there is considerable uncertainty in calculating carbon emissions due to the complex of emission pathways. In particular, the method of calculating carbon dioxide (CO2) net atmospheric flux (NAF) based on a simple gas exchange theory from sporadic data has limitations in explaining the spatiotemporal variations in the CO2 flux in stratified reservoirs. This study was aimed to analyze the spatial and temporal CO2 distribution and mass balance in Daecheong Reservoir, located in the mid-latitude monsoon climate zone, by applying a two-dimensional hydrodynamic and water quality model (CE-QUAL-W2). Simulation results showed that the Daecheong Reservoir is a heterotrophic system in which CO2 is supersaturated as a whole and releases CO2 to the atmosphere. Spatially, CO2 emissions were greater in the lacustrine zone than in the riverine and transition zones. In terms of time, CO2 emissions changed dynamically according to the temporal stratification structure of the reservoir and temporal variations of algae biomass. CO2 emissions were greater at night than during the day and were seasonally greatest in winter. The CO2 NAF calculated by the CE-QUAL-W2 model and the gas exchange theory showed a similar range, but there was a difference in the point of occurrence of the peak value. The findings provide useful information to improve the quantification of CO2 emissions from reservoirs. In order to reduce the uncertainty in the estimation of reservoir carbon emissions, more precise monitoring in time and space is required.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.413-415
• /
• 2022

Since human joints can be used as useful information for analyzing human behavior as a component of the human body, many studies have been conducted on human action recognition using joint information. However, it is a very complex problem to recognize human action that changes every moment using only each independent joint information. Therefore, an additional information extraction method to be used for learning and an algorithm that considers the current state based on the past state are needed. In this paper, we propose a human action recognition technique considering the positional relationship of connected joints and the change of the position of each joint over time. Using the pre-trained joint extraction model, position information of each joint is obtained, and bone information is extracted using the difference vector between the connected joints. In addition, a simplified neural network is constructed according to the two types of inputs, and spatio-temporal features are extracted by adding LSTM. As a result of the experiment using a dataset consisting of 9 behaviors, it was confirmed that when the action recognition accuracy was measured considering the temporal and spatial relationship features of each joint, it showed superior performance compared to the result using only single joint information.