• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.8
• /
• pp.3769-3789
• /
• 2018

Social pedestrian groups are the basic elements that constitute a crowd; therefore, detection of such groups is scientifically important for modeling social behavior, as well as practically useful for crowd video understanding. A social group refers to a cluster of members who tend to keep similar motion state for a sustained period of time. One of the main challenges of social group detection arises from the complex dynamic variations of crowd patterns. Therefore, most works model dynamic groups to analysis the crowd behavior, ignoring the existence of stationary groups in crowd scene. However, in this paper, we propose a novel unified framework for detecting social pedestrian groups in crowd videos, including dynamic and stationary pedestrian groups, based on spatiotemporal-oriented energy measurements. Dynamic pedestrian groups are hierarchically clustered based on energy flow similarities and trajectory motion correlations between the atomic groups extracted from principal spatiotemporal-oriented energies. Furthermore, the probability distribution of static spatiotemporal-oriented energies is modeled to detect stationary pedestrian groups. Extensive experiments on challenging datasets demonstrate that our method can achieve superior results for social pedestrian group detection and crowd video classification.

• The Journal of Korea Robotics Society
• /
• v.17 no.2
• /
• pp.124-132
• /
• 2022

With growth in intelligence of mobile robots, interaction with humans is emerging as a very important issue for mobile robots and the pedestrian tracking technique following the designated person is adopted in many cases in a way that interacts with humans. Among the existing multi-object tracking techniques for pedestrian tracking, Simple Online and Realtime Tracking (SORT) is suitable for small mobile robots that require real-time processing while having limited computational performance. However, SORT fails to reflect changes in object detection values caused by the movement of the mobile robot, resulting in poor tracking performance. In order to solve this performance degradation, this paper proposes a more stable pedestrian tracking algorithm by correcting object tracking errors caused by robot movement in real time using wheel odometry information of a mobile robot and dynamically managing the survival period of the tracker that tracks the object. In addition, the experimental results show that the proposed methodology using data collected from actual mobile robots maintains real-time and has improved tracking accuracy with resistance to the movement of the mobile robot.

• Journal of Korean Society of Transportation
• /
• v.34 no.2
• /
• pp.168-179
• /
• 2016

This study describes an improved model for estimating pedestrian LOS (Level of Service) by utilizing the space occupied by pedestrians. The method introduced the concept of conflict along the bi-directional pedestrian flow which enables calculating conflict area and average travel time in walking. Especially, the method incorporates the idea of generalized density concept which can consider effective walking area and pedestrian flow rates that might vary during the analysis period. After establishing methodology, adjustments of pedestrian LOS criteria in term of walking space occupied by pedestrians were performed. As a result, walking-occupied space at capacity level is 0.68 and corresponding pedestrian flow rate was calculated as 80 persons/min/m, while different pedestrian-occupied spaces were ordered to classify LOS at the points where the gradient changes. Furthermore, the statistical verification of service levels has shown that there is significant difference among all LOS categories at 5% significance level.

• Journal of Korean Society of Transportation
• /
• v.34 no.2
• /
• pp.191-206
• /
• 2016

Pedestrian performance measurement has generally used the average values of pedestrian volumes and speeds from a sample, although each individual's pedestrian behavior of the sample is mainly affected by his/her socio-economic and physical characteristics. In particular, if such individual characteristics of the sample have wide variations, the average values are hard to be representative. In this respect, this study found that a platoon of pedestrians had a major effect on their pedestrian flows on sidewalks. The more platoons was mixed on pedestrian roads, the lower pedestrian flow rate was. The indexes such as a pedestrian flow rate and an occupancy space are typically calculated by pedestrian volumes in a given period of time regardless of the existence of a platoon. However, it is plausible that a pedestrian speed and a level of pedestrian congestion can be different under the equal amount of pedestrian volumes. Therefore, this study suggested criteria for determining level of service for pedestrian considering platoon effects and confirmed that two places had different pedestrian LOSs because of the platoon effects, even if they showed the same pedestrian flow rate.

• Journal of Software Assessment and Valuation
• /
• v.17 no.2
• /
• pp.83-89
• /
• 2021

In this paper, we propose an algorithm that provide services to consumers who want to conduct business by scientifically and systematically analyzing the number of pedestrians in a specific area over a specific period of time. In this paper, we proposed the algorithm to analyze the commercial area using the pedestrian-detect algorithm in the particular region using YOLO, one of the deep learning techniques. And with one image per minute in the images, the number of pedestrians is identified and this information is used for the analysis of business district on interesting area and time, systematically and objectively.

• Journal of Trauma and Injury
• /
• v.35 no.4
• /
• pp.240-247
• /
• 2022

Purpose: Studies on the relationship between obesity and injuries, especially those sustained in pedestrian traffic accidents, are lacking. We aimed to assess the effects of obesity on the severity of injury at the time of admission to the emergency room in patients who experienced pedestrian traffic accidents. Methods: This study included trauma patients registered in the Korean Trauma Database from July 1, 2018 to December 31, 2020, whose mechanism of injury was pedestrian traffic accidents and who were treated at a single institution. Those aged below 15 years were excluded. Patients were assigned to nonobese and obese groups based on a body mass index of 25 kg/m². An Injury Severity Score of 25 or greater was considered to indicate a critical injury. Results: In total, 679 cases of pedestrian traffic accidents were registered during the study period, and 543 patients were included in the final analysis. Of them, 360 patients (66.3%) and 183 patients (33.7%) were categorized as nonobese and obese, respectively. The median age was significantly higher in the nonobese group than in the obese group (60 vs. 58 years). Multivariate analysis demonstrated that the odds ratio for critical injury in obese patients was 1.59 (95% confidence interval, 1.01-2.48) compared with nonobese patients. Conclusions: Obesity affected the likelihood of sustaining severe injuries in pedestrian traffic accidents. Future studies should analyze the effects of body mass index on the pattern and severity of injuries in patients with more diverse injury mechanisms using large-scale data.

• The Journal of Korea Robotics Society
• /
• v.15 no.3
• /
• pp.286-292
• /
• 2020

This paper is about a fall inducement system for guiding like a real fall. Reliable fall data can be used as an essential element in developing effective fall protection devices. We can get this data if the induced fall is very realistic. The proposed system analyzes gait characteristics and determines when to fall based on the pedestrian's biometric data. To estimate the fall inducement time, an active estimation algorithm was proposed using different biometric values for each pedestrian. The proposed algorithm is designed to response actively to the ratio of gait cycle and a stance period. To verify this system, an experimental environment was implemented using a multi-rail treadmill equipped with a ground reaction force measurement device. An experiment was conducted to induce falls to pedestrians using a fall inducement system. By comparing the experimental scene to the video of the actual fall, it has been confirmed that the proposed system can induce a reliable fall.

• Journal of the Korean Regional Science Association
• /
• v.38 no.2
• /
• pp.59-72
• /
• 2022

In line with increasing interest of climate change and pedestrian rights, the city of Seoul has been implementing car-free pedestrian street policies. The government and the private sector agreed to introduce carfree streets, expecting to increase the urban vibrancy and attract customers in the streets. However, studies that empirically analyze the effects of the policy are insufficient, and there are no studies that reevaluate the car-free pedestrian street policy after the COVID-19 outbreak. Therefore, this study evaluates the car-free pedestrian street policy at the commercial street, the Jongno 52-gil, using the synthetic control method. As a result of the analysis, it was found that the car-free pedestrian street policy was valid for enhancing urban vibrancy. Also, it was found that proper time interval is required to assess the policy since the peak increase of policy was observed in 3 months after the policy implementation. However, less than a year, the increase of the de facto population density on the Jongno 52-gil was revealed to be insignificant. In addition, the Jongno 52-gil was more vulnerable in the decrease of the de facto population density in the COVID-19 period than the synthetic control region. The results of this study suggest that long-term population inflow measures and flexible counterplan to contagious diseases are necessary in the car-free pedestrian street policy.

• Proceeding of Spring/Autumn Annual Conference of KHA
• /
• 2009.04a
• /
• pp.327-331
• /
• 2009

Covering structures and elevated road of Cheonggyecheon are the by-products of the developing era when functionality and efficiency were emphasized. Historic river restoration project around Cheonggyecheon which had been emerged as a threat to the safety of the citizens due to its corrosion and structural defects as time passes had been performed during the period from July, 2003 through October, 2005 under the slogan of future-oriented urban environment creation. By dint of this river restoration project, its value is being heightened as almost all the rivers of Seoul including Seongbukcheon, Chungreungcheon and Honghaecheon have been arranged by inter-active linkage. In addition, this river is becoming a place frequented by the numerous visitors at home and aborad by providing the citizens with an attractive space for utilizing leisure time. Currently in the diversified areas, an academic research on the Cheonggyecheon restoration is under progress. In this study, sampled questionnaire for the visitors was carried out by site visit with the necessity of users' satisfaction survey in terms of revitalization of pedestrian space of the city street. As a result of this, natural restoration and provision of leisure space were surveyed to be highly assessed in terms of satisfaction and insufficient provision of convenience facilities including toilet and parking space was pointed out as a matter to be improved. As Cheonggyecheon is invigorated from the obstacle of urban environment through its epochal restoration, its value provided to us will be evaluated and by supplementing the deduced problems, we intend to find its meaning in our hope that this would become a sustainable future assets providing a basic data.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1249-1252
• /
• 2022

In this paper, we propose a conceptual framework that identifies video frames in motion containing the movement of people and vehicles in traffic videos. The automatic selection of video frames in motion is an important topic in security and surveillance video because the number of videos to be monitored simultaneously is simply too large due to limited human resources. The conventional method to identify the areas in motion is to compute the differences over consecutive video frames, which has been costly because of its high computational complexity. In this paper, we reduced the overall complexity by examining only the keyframes (or I-frames). The basic assumption is that the time period between I-frames is rather shorter (e.g., 1/10 ~ 3 secs) than the usual length of objects in motion in video (i.e., pedestrian walking, automobile passing, etc.). The proposed method estimates the possibility of videos containing motion between I-frames by evaluating the difference of consecutive I-frames with the long-time statistics of the previously decoded I-frames of the same video. The experimental results showed that the proposed method showed more than 80% accuracy in short surveillance videos obtained from different locations while keeping the computational complexity as low as 20 % compared to the HM decoder.