• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.12
• /
• pp.561-568
• /
• 2021

In recent years, autonomous driving technologies have become a high-value-added technology that attracts attention in the fields of science and industry. For smooth Self-driving, it is necessary to accurately detect an object and estimate its movement speed in real time. CNN-based deep learning algorithms and conventional dense optical flows have a large consumption time, making it difficult to detect objects and estimate its movement speed in real time. In this paper, using a single camera image, fast object detection was performed using the YOLOv5 algorithm, a deep learning algorithm, and fast estimation of the speed of the object was performed by using a local dense optical flow modified from the existing dense optical flow based on the detected object. Based on this algorithm, we present a system that can predict the collision time and probability, and through this system, we intend to contribute to prevent tram accidents.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.1
• /
• pp.61-68
• /
• 2024

Despite the continuous efforts to mitigate pedestrian accidents at crosswalks, the problem persist. Vulnerable groups, including the elderly and disabled individuals are at a risk of being involved in traffic incidents. This paper proposes the implementation of object detection algorithm using the YOLO v5 model specifically for pedestrians using assistive devices like wheelchairs and crutches. For this research, data was collected and utilized through image crawling, Roboflow, and Mobility Aids datasets, which comprise of wheelchair users, crutch users, and pedestrians. Data augmentation techniques were applied to improve the model's generalization performance. Additionally, ensemble techniques were utilized to mitigate type 2 errors, resulting in 96% recall rate. This demonstrates that employing ensemble methods with a single YOLO model to target transportation-disadvantaged individuals can yield accurate detection performance without overlooking crucial objects.

• PNF and Movement
• /
• v.22 no.1
• /
• pp.43-54
• /
• 2024

Purpose: Falls among persons older than 65 years are a significant concern due to their frequency and severity. This study aimed to develop a vest-type embedded artificial intelligence (AI) system capable of detecting and predicting falls in various scenarios. Methods: In this study, we established and developed a vest-type embedded AI system to judge and predict falls in various directions and situations. To train the AI, we collected data using acceleration and gyroscope values from a six-axis sensor attached to the seventh cervical and the second sacral vertebrae of the user, considering accurate motion analysis of the human body. The model was constructed using a neural network-based AI prediction algorithm to anticipate the direction of falls using the collected pedestrian data. Results: We focused on developing a lightweight and efficient fall prediction model for integration into an embedded AI algorithm system, ensuring real-time network optimization. Our results showed that the accuracy of fall occurrence and direction prediction using the trained fall prediction model was 89.0% and 78.8%, respectively. Furthermore, the fall occurrence and direction prediction accuracy of the model quantized for embedded porting was 87.0 % and 75.5 %, respectively. Conclusion: The developed fall detection and prediction system, designed as a vest-type with an embedded AI algorithm, offers the potential to provide real-time feedback to pedestrians in clinical settings and proactively prepare for accidents.

• Journal of Korean Society of Disaster and Security
• /
• v.16 no.4
• /
• pp.137-145
• /
• 2023

Accidents involving deaths of children and pedestrians continue to occur in child protection zones, also known as school zones. Over the three years from 2019 to 2021, 11 child deaths occurred in school zones in Korea. In December 2022, a fatal accident occurred among an elementary school student in a school zone near an elementary school in Gangnam-gu, Seoul. In April 2023, an accident occurred in which an elementary school student died in a school zone near Seo-gu Middle School in Daejeon and an elementary school in Yeongdo-gu, Busan. Until recently, child deaths continued to occur on sidewalks in child protection zones. It is time to closely review the status of safety facilities installed in school zones and related regulations. We investigated and analyzed the current status of safety facilities and related regulations installed on sidewalks in child protection zones in Korea, and identified matters necessary for institutional improvement regarding safety facilities for sidewalks in child protection zones.

• Journal of Korean Society of Disaster and Security
• /
• v.6 no.1
• /
• pp.79-86
• /
• 2013

As car-oriented road policies have been made forward so far, relatively pedestrians' walking conditions are so in poor environments that more than two thousand pedestrians die from car accidents every year. Pedestrians' walking right has been severely invaded like that. Pedestrians' walking right is a right that people are able to walk safely and comfortably in pleasant surroundings as long as they don't threaten the public safety, order maintenance, and welfare. The government has an obligation to provide safe, comfortable, and pleasant environments to pedestrians. Recently interests in pedestrians' safety are increasing, government-driven supports have been made to make safe, pleasant, and healthy walking surroundings. As poor walking condition improvement projects cost high, they should be progressed to accomplish maximal effects using finite finances efficiently, and post feasibility evaluations of the projects should be severely estimated. However site selecting indicators which satisfy with the goal for composing safe working surroundings have not been decided yet, though currently it has a legal basis to specify walking condition improvement sites by the Law for Pedestrians' safety and Comfort Increasement. Therefore this study focuses on suggesting improved ways for selecting sites where pedestrians' safe environment project by reviewing previous research. When project sites are selected, evaluation indicators related to awareness survey of residents and history should be excluded, and disaster safety assessments for walking safety facilities, latent human hazards and natural disasters like a strong wind are proposed besides evaluations on pedestrians' safety and walking environment for matching with the purpose of the project to make safe working surroundings.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.10 no.4
• /
• pp.107-115
• /
• 2011

Since 2009 The Korean National Police Agency has been pushed "permissive left-turn" forward in order to making progress in Korean traffic operations system. Preceding researches manuals and guidelines present 6 standards(# of accidents # of lanes, pedestrian volume sight # of permissive left-turn lanes vehicle volume) as installation permissive left-turn. But in practical affairs it is most important that secure enough Gap-time between permissive left-turn vehicle and opposite through lane vehicle to make permissive left-turn vehicle move safer and more efficiently. This study suggests applying gap acceptance theory in microscopic model to permissive left-turn installation standards. Analysis methods of this study are field data survey statistical analysis and microscopic simulation analysis. This study collected field data by using AVI recording and measured permissive left-turn vehicle intersection passing time(T1) and against the opposite through lane vehicle Gap-time(T2). And statistical analysis performed about two values that measured before to predict the functionality between T1 and T2. These studies to overcome the limit of sample size carried out a microscopic simulation(VISSIM) plan and collect more samples to input statistical analysis.

• Journal of Korean Society of Transportation
• /
• v.21 no.4
• /
• pp.133-144
• /
• 2003

The current procedure of estimating accident reduction benefit shows fixed accident rates for each level of roads without considering the various characteristics of roadway geometries, and traffics. In this study, in order to solve the problems mentioned in the above, models were developed considering the characteristics of roadway alignments and traffic characteristics. The developed models can be used to estimate the accident rates on new or improved roads, In this study, only urban highways were included as a beginning stage. First of all. factors influencing accident rates were selected. Those factors such as traffic volumes. number of signalized intersections, the number of connecting roads, number of pedestrian traffic signals, existence of median barrier, and the number of road lane are also selected based upon the obtainability at the planning stage of roads. The relationship between the selected factors and accident rates shows strong correlation statistically. In this study, roads were classified into 4 groups based on number of lanes, level of roads and the existence of median barriers. The regression analysis had been performed for each group with actual data associated with traffic, roads. and accidents. The developed regression models were verified with another data set. In this study, in order to develop the proposed models, only data on a limited area were used. In order to represent whole area of the country with the developed models. the models should be re-analyzed with vast data.

• Journal of Trauma and Injury
• /
• v.21 no.1
• /
• pp.22-27
• /
• 2008

Purpose: Treatment and prognosis in patients with pelvic bone fracture depend on the characteristics of the fracture and the stability of the pelvic ring. The purpose of this study is to analyze the characteristics of and the relationships between fracture patterns, injury mechanisms, clinical courses, and prognoses according to the hemodynamic pattern. Methods: Between January 2004 and September 2006, 89 patients under diagnosis of pelvic bone fracture were retrospectively analyzed on the basis of medical records and radiologic examinations. Patients with confirmed hemorragic shock with a systolic pressure of less than 90 mmHg were defined as the shock group. Young's classification was used to characterize fracture patterns. Factors relating to the clinical manifestation and to treatments such as transfusion and surgery were analytically compared. Results: The mean age of the patients was $48.8{\pm}18.7$, among which 49 (55.1%) were male. The numbers of shock and non-shock patients were 35 (39.3%) and 54 (60.7%) respectively. Eighteen (51.4%) of the shock patient had injuries resulting from pedestrian accidents (p=0.008). According to Young's classification, lateral impact fractures amounted to 20 and 33, front-rear impact fractures to 9 and 20, and multiple fractures to 6 and 1 among the shock and non-shock patients, respectively (p=0.027). Thirty-nine (39) cases in non-shock injuries were conservatively managed while 18 cases in shock injuries were surgically treated. In the shock group, the liver and the kidney were often damaged, as well. Among the shock patients, the average admission period was $7.5{\pm}8.7days$ in intensive care and $55.1{\pm}47.9days$ in total, which were longer than the corresponding numbers of days for the non-shock patients (p<0.05). No deaths occurred in the non-shock group while 5 deaths (14.2%) occurred in the shock group (p=0.007). Conclusion: The factors affecting hemodynamic instability in patients with pelvic bone fracture are injury mechanism, classification of fracture, and associated injuries.

• Journal of Trauma and Injury
• /
• v.35 no.3
• /
• pp.195-201
• /
• 2022

Purpose: The objective of this study was to determine the effects of coronavirus disease 2019 (COVID-19) on the volume of trauma patients, the number of orthopedic trauma operations, and the severity of injuries. We also investigated the correlations between social distancing and these variables. Methods: This was a retrospective review of trauma patient cases at a single focused training center for trauma in Korea from January 2017 to April 2021. The COVID-19 group included patients treated from January 1 to April 30 in 2020 and 2021, and the control group included patients treated during the same months from 2017 to 2019. The volume of trauma patients according to the level of social distancing was evaluated among patients treated from August 2, 2020 to November 23, 2020. Results: The study included 3,032 patients who presented to the emergency department with traumatic injuries from January to April 2017 to 2021. The average number of patients was 646.7 and 546.0 in the control and COVID-19 groups, respectively. The percentage of patients injured in traffic accidents (TAs) decreased from 25.0% to 18.2% (P<0.0001). The proportions of in-car TAs and pedestrian TAs also decreased from 6.7% and 10.8% to 3.5% and 6.0%, respectively (P=0.0002 and P<0.0001). The percentage of bicycle TAs increased from 2.4% to 4.0% (P=0.0128). The proportion of patients with an Injury Severity Score above 15 and the mortality rate did not change significantly. As the level of social distancing increased, the number of trauma patients and the number of trauma injuries from TAs decreased. The number of orthopedic trauma operations also depended on the social distancing level. Conclusions: The number of trauma patients presenting to the emergency department decreased during the COVID-19 period. The volume of trauma patients and orthopedic trauma operations decreased as the social distance level increased.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.22 no.2
• /
• pp.164-181
• /
• 2023

In a low speed limit environment, the speed limit tolerance of automated traffic enforcement devices is very high, which is one of the main factors for the low compliance rate. Therefore, in this study, we aimed to the improve the speed limit tolerance and to present a new standard. The effects of the operator and user errors that can cause speeding by drivers were analyzed. Based on the results of the analysis, an improvement of the tolerance was proposed by applying an error in the enforcement device and GPS speed. In addition, long-term expected safety effects such as the accident rate and severity were estimated from the operator's perspective when improving the tolerance. As a result of the estimation, the speed limit compliance rate, accident rate, and change rate of a number of severe accidents due to speed change, and pedestrian traffic accident mortality rate were all improved in all speed limit environments. The introduction of the proposed improvement is expected to improve road safety significantly.