• Journal of Power System Engineering
• /
• v.13 no.5
• /
• pp.76-81
• /
• 2009

The fatalities of pedestrian account for about 21.2% of all fatalities at 2007 year in Korea. To reconstruct exactly the accident, it is important to calculate the throw distance of pedestrian in car to pedestrian accident. The frontal shape of SUV vehicle is dissimilar to passenger car and bus, so the trajectory and throw distance of pedestrian by SUV vehicle is not the same of passenger car and bus. The influencing on it can be classified into the factors of vehicle and pedestrian, and road factor. It was analyzed by PC-CRASH for simulation, and SPSS s/w was used for regression analysis. From the simulation results, the maximum impact energy of multi-body of pedestrian was occurred to that of torso body at the same time. And the throw distance increased with the increasing of impact velocity, and decreased with the increasing of impact offset. Also it decreased with the increasing of velocity of pedestrian at accident, and the throw distance of wet road was longer than that of dry road. Finally, the regression analysis model of SUV(Nissan Pathfinder type)vehicle in car to pedestrian accident was as follows; $$disti_i=-0.87-0.11offseti_i+0.69speed_i-4.27height_i+0.004walk_i+0.63wet_i+{\epsilon}_i$$.

• Journal of the Korean Society of Safety
• /
• v.30 no.1
• /
• pp.94-103
• /
• 2015

This study analyzes the influence factors on elderly pedestrian accident. Elderly people are easy to be badly injured by car accidents compared to younger people. Therefore, various plans and measures are required to protect elderly pedestrian from accidents. However, pedestrian accidents studies only focused on microscopic factors such as attribute of driver, pedestrian, road design. In order to prevent pedestrian accident and reduce the severity of the accident, not only microscopic factors but macroscopic variables such as urban planning and facility should be considered. In this regard, this study develops an ordered probit model introduced the characteristics of urban facility which were not considered in the previous studies. The result shows that there is higher level of accident severity in such areas as large commercial area, well-developed area with transportation infrastructure service and non-pedestrian safety zone. Thus, various and appropriate countermeasures should be prepared in order that pedestrian accident can be prevented in the areas mentioned above. In addition to the aforementioned variables, it is revealed that other variables including vehicle speed, gender and age of pedestrian, weather condition, type of vehicle, etc. partly affect the severity of pedestrian accident.

• Journal of the Semiconductor & Display Technology
• /
• v.22 no.2
• /
• pp.128-132
• /
• 2023

Recently, with the development of deep learning technology, research on pedestrian attribute recognition technology using deep neural networks has been actively conducted. Existing pedestrian attribute recognition techniques can be obtained in such a way as global-based, regional-area-based, visual attention-based, sequential prediction-based, and newly designed loss function-based, depending on how pedestrian attributes are detected. It is known that the performance of these pedestrian attribute recognition technologies varies greatly depending on the type of backbone network that constitutes the deep neural networks model. Therefore, in this paper, several backbone networks are applied to the baseline pedestrian attribute recognition model and the performance changes of the model are analyzed. In this paper, the analysis is conducted using Resnet34, Resnet50, Resnet101, Swin-tiny, and Swinv2-tiny, which are representative backbone networks used in the fields of image classification, object detection, etc. Furthermore, this paper analyzes the change in time complexity when inferencing each backbone network using a CPU and a GPU.

• Journal of Auto-vehicle Safety Association
• /
• v.9 no.4
• /
• pp.14-19
• /
• 2017

The mortality rate of car-pedestrian accidents is quite high compared to the frequency of accident. Recently, governments and insurance companies tend to establish and implement new safety standards for pedestrian protection such as EURO-NCAP and K-NCAP. The performance for the pedestrian protection has been gradually improved, but it is still insufficient. Therefore, various studies for the pedestrian protection are being carried out. The car-pedestrian accident is simulated in order to study to the upper legform test of the EURO-NCAP protocol. A pedestrian dummy model is employed and the results are discussed.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.19 no.5
• /
• pp.82-91
• /
• 2011

In this paper, a planar model for mechanics of a vehicle/pedestrian collision incorporating road gradient is derived to evaluate the pre-collision speed of vehicle. It takes into account a few physical variables and parameters of popular wrap and forward projection collisions, which include horizontal distance traveled between primary and secondary impacts with the vehicle, launch angle, center-of-gravity height at launch, distance from launch to rest, pedestrian-ground drag factor, the pre-collision vehicle speed and road gradient. The model including road gradient is derived analytically for reconstruction of pedestrian collision accidents, and evaluates the vehicle speed from the pedestrian throw distance. The model coefficients have physical interpretations and are determined through direct calculation. This work shows that the road gradient has a significant effect on the evaluation of the vehicle speed and must be considered in accident cases with inclined road. In additions, foreign/domestic empirical cases and multibody dynamic simulation results are used to construct a least-squares fitted model that has the same structure of the analytical one that provides an estimate of the vehicle speed based on the pedestrian throw distance and the band within which the vehicle speed would be expected to be in 95% of cases.

• Spatial Information Research
• /
• v.22 no.2
• /
• pp.53-62
• /
• 2014

The floor field model is the micro pedestrian model based on a cellular automata for modeling pedestrian movement in the interior space using the static and dynamic floor field. It regards a form of pedestrian as square but the actual pedestrian's shape and size are similar to ellipsoid or rectangle. Because of this, we are difficult to give a rotation effect to pedestrians and there is a limit to reflect an impact of clogging and jamming. Also, this model is not able to reflect an impact of a posture and visibility effectively in the pedestrian movement. In this study, we suggest the improved pedestrian model incorporating the actual shape and size of pedestrian. The pedestrian's shape is defined not square but rectangle which is close to the actual body size of Korean. Also, we define the model which is able to represent the impact of clogging and jamming between pedestrians by adding the pedestrian's posture. We develop the simulator for testing the suggested model and study the difference between two models by comparing a number of effects. As a result, we could confirm solving the problem with dynamic value in the existed model and reflecting the panic effect in evacuation situation.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.36 no.2
• /
• pp.259-268
• /
• 2016

The core strategy of transit oriented development (TOD) is to promote high density mixed land use around railway stations. Case studies in advanced countries show that provision of policies for comprehensive maintenance of pedestrian facilities around railway station spheres is being pursued with efficacy. In spite of the importance placed on integrated pedestrian maintenance, domestic construction of integrated pedestrian infrastructure around railway station spheres lacks direction. Thus, there is a clear need for an evaluation standard that can provide the foundation for judgments on TOD improvement. This research proposes a network model that consolidates the interior of the station as well as its surrounding areas to determine the ease of pedestrian flow for effective TOD evaluation. The model considers the railway station and surrounding areas as an assembled network of pedestrian flow. The path chosen by the pedestrian is defined as the optimal degree of inconvenience, and expands on Wardrop's User Equilibrium (1952). To assess the various circumstances that arise on pedestrian facilities including congestion of the pedestrian pathway, constrained elevator capacity, and wait at the crosswalk, a variational inequality based pedestrian equilibrium distribution model is introduced.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.7
• /
• pp.2321-2338
• /
• 2021

Compared with vehicle trajectories, pedestrian trajectories have stronger degrees of freedom and complexity, which poses a higher challenge to trajectory prediction tasks. This paper designs a mode to divide the trajectory of pedestrians at a traffic intersection, which converts the trajectory regression problem into a trajectory classification problem. This paper builds a deep model for pedestrian trajectory prediction at intersections for the task of pedestrian short-term trajectory prediction. The model calculates the spatial correlation and temporal dependence of the trajectory. More importantly, it captures the interactive features among pedestrians through the Attention mechanism. In order to improve the training speed, the model is composed of pure convolutional networks. This design overcomes the single-step calculation mode of the traditional recurrent neural network. The experiment uses Vulnerable Road Users trajectory dataset for related modeling and evaluation work. Compared with the existing models of pedestrian trajectory prediction, the model proposed in this paper has advantages in terms of evaluation indicators, training speed and the number of model parameters.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.3
• /
• pp.1464-1480
• /
• 2019

Robust tracking of infrared (IR) pedestrian targets with various backgrounds, e.g. appearance changes, illumination variations, and background disturbances, is a great challenge in the infrared image processing field. In the paper, we address a new tracking method for IR pedestrian targets via multi-feature local sparse representation (SR), which consists of three important modules. In the first module, a multi-feature local SR model is constructed. Considering the characterization of infrared pedestrian targets, the gray and edge features are first extracted from all target templates, and then fused into the model learning process. In the second module, an effective tracker is proposed via the learned model. To improve the computational efficiency, a sliding window mechanism with multiple scales is first used to scan the current frame to sample the target candidates. Then, the candidates are recognized via sparse reconstruction residual analysis. In the third module, an adaptive dictionary update approach is designed to further improve the tracking performance. The results demonstrate that our method outperforms several classical methods for infrared pedestrian tracking.

• Journal of IKEEE
• /
• v.17 no.4
• /
• pp.469-477
• /
• 2013

This paper enhances the speed of a pedestrian detection using an estimation of feature information based on integral image. Pedestrian model or input image should be resized to the size of various pedestrians. In case that the size of pedestrian model would be changed, pedestrian models with respect to the size of pedestrians should be required. Reducing the size of pedestrian model, however, deteriorates the quality of the model information. Since various features according to the size of pedestrian models should be extracted, repetitive feature extractions spend the most time in overall process of pedestrian detection. In order to enhance the processing time of feature extraction, this paper proposes the fast extraction of pedestrian features based on the estimate of integral image. The efficiency of the proposed method is evaluated by comparative experiments with the Channel Feature and Adaboost training using INRIA person dataset.