• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.9
• /
• pp.1244-1253
• /
• 2021

In this paper, we propose a method for predicting in advance whether pedestrians will enter the hazardous area after the current time using the pedestrian trajectory prediction method and an efficient simplification method of the trajectory prediction network. In addition, we propose a method to apply KD(Knowledge Distillation) to a small network for real-time operation in an embedded environment. Using the correlation between predicted future paths and hazard zones, we determined whether to enter or not, and applied efficient KD when learning small networks to minimize performance degradation. Experimentally, it was confirmed that the model applied with the simplification method proposed improved the speed by 37.49% compared to the existing model, but led to a slight decrease in accuracy. As a result of learning a small network with an initial accuracy of 91.43% using KD, It was confirmed that it has improved accuracy of 94.76%.

• Journal of the Architectural Institute of Korea Planning & Design
• /
• v.35 no.1
• /
• pp.47-54
• /
• 2019

This study proposes a new ERAM model with building population weighting. Previous studies of applying weightings on ERAM model on the scale of urban space were focused on the relationship between the street and the human behavior. However, this study focuses on the influences that buildings give to human behavior and develops a building population weighted ERAM model. This research starts by analyzing ERAM model to its basic compositions, which are adjacency matrix and row vector. It applies building population weighting to the row vector, while previous studies put weightings in the adjacency matrix. Building population weighted ERAM model calculates the building population weighting based on GIS data, which provides objective and massive data of buildings in the urban scale. For the verification of the model, Insa-dong and Myeong-dong were analyzed with both ERAM model and building population weighted ERAM model. The results were analyzed through the correlation test with actual pedestrian population data of the two districts. As a result, the explanation ability of building population weighted ERAM model for the pedestrian population turned out to be higher than the ERAM model. Since building population weighted ERAM model has the structure that can be combined with other weighted ERAM models, it is expected to develop a multi-weighted ERAM model with better explanation ability as a further study.

• Journal of Environmental Impact Assessment
• /
• v.26 no.4
• /
• pp.242-256
• /
• 2017

In this paper, dispersion scenarios concerning various meteorological conditions and real urban structures were made to estimate the impacts of hazardous substance leakage accidents and to reduce damages. Based on the scenario of the hazardous substance dispersion, the characteristics of the risk in the pedestrian environment were analyzed in Gangnam, Seoul. The scenarios are composed of 48 cases according to the meteorological conditions of wind direction and wind speed. In order to analyze the dispersion characteristics of the hazardous substances, simulations were conducted using a computational fluid dynamic (CFD) model with hydrogen fluoride releases. The validation for the simulated wind was conducted at a specific period, and all the calculated verification indices were within the valid range. As a result of simulated dispersion field at pedestrian level, it was found that the dispersion pattern was influenced by the flow, which was affected by the artificial obstacles. Also, in the case of the weakest wind speed of the inflow, the dispersion of the hazardous substance appeared in the direction of the windward side at the pedestrian level due to the reverse flow occurred at lower layers. Through this study, it can be seen that the artificial structures forming the city have a major impact on the flow formed in urban areas. The proposed approach can be used to simulate the dispersion of the hazardous substances and to assess the risk to pedestrians in the industrial complexes dealing with actual hazardous substances in the future.

• Journal of the Korean Regional Science Association
• /
• v.36 no.2
• /
• pp.25-36
• /
• 2020

In recent urban policies, the importance of 'health' has increased significantly. In Korea, the deterioration of physical health due to environmental problems caused by rapid urbanization and the deterioration of mental health due to the disconnection of social relations in a competitive society are constantly occurring. this study aims to examine the effect of the satisfaction and safety recognition on pedestrian environment on the subjective health status focused on gender difference of the intermediating effect of social capital. I used 2018 Seoul Survey and employed Structural Equation Model multi-group analysis. According to the empirical analysis, in male groups, the satisfaction and safety recognition on the pedestrian environment affect social capital. Also, satisfaction and social capital affect subjective health status. On the other hand, in female group, perception of pedestrian environment affect social capital. In addition, satisfaction, safety recognition and social capital affect subjective health status. Therefore, interdisciplinary study in the field of health and urban policy and discriminatory policy for strategic urban policy for citizen's health is necessary.

• Journal of Korea Multimedia Society
• /
• v.18 no.11
• /
• pp.1289-1301
• /
• 2015

Pedestrian detection (PD) is an essential task in various applications and sliding window-based methods utilizing HOG (Histogram of Oriented Gradients) or HOG-like descriptors have been shown to be very effective for accurate PD. However, due to exhaustive search across images, PD methods based on sliding window usually require heavy computational time. In this paper, we propose a real-time PD method for embedded visual surveillance with fixed backgrounds. The proposed PD method employs HOG descriptors as many PD methods does, but utilizes selective search so that it can save processing time significantly. The proposed selective search is guided by restricting searching to candidate regions extracted from Adaptive Gaussian Mixture Model (AGMM)-based background subtraction technique. Moreover, approximate computation of HOG descriptor and implementation in fixed-point arithmetic mode contributes to reduction of processing time further. Possible accuracy degradation due to approximate computation is compensated by applying an appropriate one among three offline trained SVM classifiers according to sizes of candidate regions. The experimental results show that the proposed PD method significantly improves processing speed without noticeable accuracy degradation compared to the original HOG-based PD and HOG with cascade SVM so that it is a suitable real-time PD implementation for embedded surveillance systems.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.20 no.2
• /
• pp.64-69
• /
• 2012

Head injury is one of the most common cause of deaths in car-to-pedestrian collisions. To reduce the severity of such injuries, many international safety committees have performed headform impact test for pedestrian protection. In this paper, an adult headform impactor model is developed based on the finite element (FE) method and validated through the numerical simulation. The skin material of headform impactor is known as polyvinyl chloride skin (PVC) and its material was assumed as viscoelastic. The viscoelastic parameters of headform skin are identified by a series of trial and error methods. The new developed FE adult headform impactor is verified by the drop test and FE JARI adult headform impactor provided by Madymo program.

• Wind and Structures
• /
• v.11 no.1
• /
• pp.51-70
• /
• 2008

CFD is applied to evaluate pedestrian wind comfort at outdoor platforms in a high-rise apartment building. Model validation is focused on generic building sub-configurations that are obtained by decomposition of the actual complex building geometry. The comfort study is performed during the design stage, which allows structural design changes to be made for wind comfort improvement. Preliminary simulations are performed to determine the effect of different design modifications. A full wind comfort assessment study is conducted for the final design. Structural remedial measures for this building, aimed at reducing pressure short-circuiting, appear to be successful in bringing the discomfort probability estimates down to acceptable levels. Finally, the importance of one of the main sources of uncertainty in this type of wind comfort studies is illustrated. It is shown that the uncertainty about the terrain roughness classification can strongly influence the outcome of wind comfort studies and can lead to wrong decisions. This problem is present to the same extent in both wind tunnel and CFD wind comfort studies when applying the same particular procedure for terrain relation contributions as used in this paper.

• Journal of KIISE:Software and Applications
• /
• v.33 no.5
• /
• pp.499-507
• /
• 2006

Recently human gait has been considered as a useful biometric supporting high performance human identification systems. This paper proposes a view-based pedestrian identification method using the dynamic silhouettes of a human body modeled with the Hidden Markov Model(HMM). Two types of gait models have been developed both with an endless cycle architecture: one is a discrete HMM method using a self-organizing map-based VQ codebook and the other is a continuous HMM method using feature vectors transformed into a PCA space. Experimental results showed a consistent performance trend over a range of model parameters and the recognition rate up to 88.1%. Compared with other methods, the proposed models and techniques are believed to have a sufficient potential for a successful application to gait recognition.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.175-177
• /
• 2021

The parking area for the disabled is a parking facility for the pedestrian disabled and is a parking space for securing pedestrian safety passage for the disabled. However, due to the lack of social awareness of areas for the disabled, the use of parking areas is restricted, and violations such as illegal parking and obstruction of parking are increasing every year. Therefore, in this study, we propose a system to crack down on illegal parking in handicapped parking areas using the YOLOv5 model, a deep learning object recognition model to improve parking interference within parking spaces.

• Journal of Korean Society of Transportation
• /
• v.20 no.3
• /
• pp.105-113
• /
• 2002

As the number of pedestrian accident increases, the reconstruction of an accident becomes important to find the source of the fault. Generally, accidents are reconstructed by the intuition of experts or primitive physics. A reconstruction method is proposed using sophisticated optimization technology. At first, a dynamic simulation model is established for the accident environment. Occupant analysis for automobile crashworthiness is employed. The situation before an accident is identified by optimization. The impact velocity and the position of the pedestrian are utilized as design variables. The design variables are found by minimizing the difference between the simulation and the real accident. The optimization process is performed by linking an occupant analysis program MADYMO to an optimization program VisualDOC. Since the involved analysis is dynamics and highly nonlinear, response surface method is selected for the optimization process. Problems are solved for various situations.