• 대한인간공학회지
• /
• 제35권4호
• /
• pp.277-291
• /
• 2016

Objective: The aim of this study is to propose a unit touch gesture model, which would be useful to predict the performance time on mobile devices. Background: When estimating usability based on Model-based Evaluation (MBE) in interfaces, the GOMS model measured 'operators' to predict the execution time in the desktop environment. Therefore, this study used the concept of operator in GOMS for touch gestures. Since the touch gestures are comprised of possible unit touch gestures, these unit touch gestures can predict to performance time with unit touch gestures on mobile devices. Method: In order to extract unit touch gestures, manual movements of subjects were recorded in the 120 fps with pixel coordinates. Touch gestures are classified with 'out of range', 'registration', 'continuation' and 'termination' of gesture. Results: As a results, six unit touch gestures were extracted, which are hold down (H), Release (R), Slip (S), Curved-stroke (Cs), Path-stroke (Ps) and Out of range (Or). The movement time predicted by the unit touch gesture model is not significantly different from the participants' execution time. The measured six unit touch gestures can predict movement time of undefined touch gestures like user-defined gestures. Conclusion: In conclusion, touch gestures could be subdivided into six unit touch gestures. Six unit touch gestures can explain almost all the current touch gestures including user-defined gestures. So, this model provided in this study has a high predictive power. The model presented in the study could be utilized to predict the performance time of touch gestures. Application: The unit touch gestures could be simply added up to predict the performance time without measuring the performance time of a new gesture.

• International journal of advanced smart convergence
• /
• 제8권4호
• /
• pp.75-81
• /
• 2019

Though recent advancement of deep learning methods have provided satisfactory results from large data domain, somehow yield poor performance on few-shot classification tasks. In order to train a model with strong performance, i.e. deep convolutional neural network, it depends heavily on huge dataset and the labeled classes of the dataset can be extremely humongous. The cost of human annotation and scarcity of the data among the classes have drastically limited the capability of current image classification model. On the contrary, humans are excellent in terms of learning or recognizing new unseen classes with merely small set of labeled examples. Few-shot learning aims to train a classification model with limited labeled samples to recognize new classes that have neverseen during training process. In this paper, we increase the backbone depth of the embedding network in orderto learn the variation between the intra-class. By increasing the network depth of the embedding module, we are able to achieve competitive performance due to the minimized intra-class variation.

• 한국인지과학회:학술대회논문집
• /
• 한국인지과학회 2005년도 춘계학술대회
• /
• pp.9-13
• /
• 2005

A comprehension-based computational model of pilot action planning called ADAPT is presented to model pilot performance in a flight simulation context. Individual pilots were asked to execute a series of flight maneuvers using a flight simulator, and their eye-scanning, control movements, and flight performance were recorded in a time-synched database. Computational models of each of the 25 individual pilots were constructed, and the individual models simulated execution of the same flight maneuvers performed by human pilots. The time-synched eye-scanning, control movements, and flight performance of individual pilots and their respective models were compared to test ADAPT's predictive validity.

• Journal of Mechanical Science and Technology
• /
• 제19권9호
• /
• pp.1742-1752
• /
• 2005

This paper presents a Multiple Target Tracking (MTT) Adaptive Cruise Control (ACC) system which consists of three parts; a multi-model-based multi-target state estimator, a primary vehicular target determination algorithm, and a single-target adaptive cruise control algorithm. Three motion models, which are validated using simulated and experimental data, are adopted to distinguish large lateral motions from longitudinally excited motions. The improvement in the state estimation performance when using three models is verified in target tracking simulations. However, the performance and safety benefits of a multi-model-based MTT-ACC system is investigated via simulations using real driving radar sensor data. The MTT-ACC system is tested under lane changing situations to examine how much the system performance is improved when multiple models are incorporated. Simulation results show system response that is more realistic and reflective of actual human driving behavior.

• Journal of Communications and Networks
• /
• 제13권2호
• /
• pp.142-148
• /
• 2011

Because currently available wireless technologies are not appropriate for wireless body area networks (WBANs), the IEEE 802.15.6 standard was introduced by the IEEE 802.15.6 Task Group to satisfy all the requirements for a monitoring system that operates on, in, or around the human body. In this work, we develop an analytical model for evaluating the performance of an IEEE 802.15.6-based WBAN under saturation condition and a noisy channel. We employ a three-dimensional Markov chain to model the backoff procedure as specified in the standard. Probability generating functions (PGFs) are used to compute the performance descriptors of the network. The results obtained from the analytical model are validated by simulation results. Our results indicate that under saturation condition, the medium is accessed by the highest user priority nodes at the vast majority of time while the other nodes are starving.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1999년도 봄 학술발표회 논문집
• /
• pp.305-312
• /
• 1999

This study is intended to propose a systematic and practical life cycle cost(LCC) model for the development of the reliability-based seismic safety and cost-effective performance criteria for design and upgrading of long-span PC bridges. The LCC models consist of five cost functions such as initial cost, repair/replacement cost, human losses, road user cost, and indirect losses of regional economy. The proposed model Is successfully expressed in temrs of Park-Ang damage indices and life cycle damage probability obtained from SMART-DRAIN-2DX which is an existing algorithm for nonlinear time history analysis. The proposed LCC model is successfully applied to a viaduct constructed by PSM, in Seoul. Based on the observations, the proposed systematic procedure for the formulation of LCC model may be useful for the development of the reliability-based seismic safety and cost-effective performance criteria for design and upgrading of long-span PC bridges.

• 한국자동차공학회논문집
• /
• 제5권3호
• /
• pp.209-219
• /
• 1997

A fuzzy driver model based on a preview-predictor and yaw rate is developed. The model is used to investigate the handling performance of two wheel steering system(2WS) and four wheel steering system(4WS) vehicles. The two degree-of- freedom model which has yaw and lateral motion predicts the path of the vehicles. Based upon the yaw rate and lateral deviations, the fuzzy engine describes the human driver's complicated control behavior which is adjusted for the driving environment. Both typical single lane change maneuver and double lane change maneuver are adopted to demonstrate the feasibility of fuzzy driver model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권8호
• /
• pp.3136-3150
• /
• 2015

A vision-based 3D tracking of articulated human hand is one of the major issues in the applications of human computer interactions and understanding the control of robot hand. This paper presents an improved approach for tracking and recovering the 3D position and orientation of a human hand using the Kinect sensor. The basic idea of the proposed method is to solve an optimization problem that minimizes the discrepancy in 3D shape between an actual hand observed by Kinect and a hypothesized 3D hand model. Since each of the 3D hand pose has 23 degrees of freedom, the hand articulation tracking needs computational excessive burden in minimizing the 3D shape discrepancy between an observed hand and a 3D hand model. For this, we first created a 3D hand model which represents the hand with 17 different parts. Secondly, Random Forest classifier was trained on the synthetic depth images generated by animating the developed 3D hand model, which was then used for Haar-like feature-based classification rather than performing per-pixel classification. Classification results were used for estimating the joint positions for the hand skeleton. Through the experiment, we were able to prove that the proposed method showed improvement rates in hand part recognition and a performance of 20-30 fps. The results confirmed its practical use in classifying hand area and successfully tracked and recovered the 3D hand pose in a real time fashion.

• 정보처리학회논문지D
• /
• 제16D권3호
• /
• pp.395-406
• /
• 2009

현재 정부 각 부처는 “정보시스템의 효율적 도입 및 운영 등에 관한 법률”에 의해 EA를 도입하여 운용하고 있으나 이의 평가 체계 및 지표는 제대로 갖춰져 있지 않다. 특히 EA의 성과평가를 위한 모델이 없을 뿐만 아니라 측정 할 수 있는 방법도 미흡하다. 본 연구에서는 EA성과평가를 위해 국내 범정부 성과참조모델과 미 연방 성과참조모델을 기반으로 8개의 평가 영역 및 17개의 평가지표를 도출하였다. 또한, 미 OMB 및 국내 범정부 EA성숙도 모델을 기반으로 8개의 평가 영역 및 10개의 평가지표를 도출하여, 이를 기반으로 EA 성과평가를 위한 업무, 고객, 프로세스, 인적자본 및 기술의 5개 영역으로 구성된 성과평가 모델을 제안하였다. 제안한 모델의 검증을 위해 3개 기관에 적용하여 각 기관의 EA 성과를 평가하였고, 평가 결과를 검증하기 위해 범정부 EA성숙도 모델의 성과평가 영역과도 비교 평가하였다. 또한, 제안한 성과평가모델의 측정지표를 기존의 EA성숙도 모델 및 성과참조모델과도 비교 분석하였다. 이러한 비교 분석 결과를 바탕으로 제안한 성과평가모델을 검증하였다.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2010년도 춘계학술대회 논문집
• /
• pp.2048-2051
• /
• 2010

Almost all companies have paid much attention to the safety management ranging from maintenance to operation even at the stage of designing in order to prevent accidents, but fatal accidents continue to increase throughout the world. In particular, it is essential to systematically prevent such fatal accidents as fire, explosion or leakage of toxic gas at factories in order to not only protect the workers and neighbors but also prevent economic losses and environmental pollution. In addition, HRA may be used to detect the human errors which may cause accidents or trace back to any mistake on the part of workers. Usually, HRA technique is used in association with other risk assessment techniques. Moreover, it can serve to enumerate the human errors which may occur during operation or down-time or correct the existing system to reduce the mistakes. This work focuses on the coincidence of human error and mechanical failure for management of human error, and on some important performance shaping factors to propose a method for improving safety effectively of the process industries.