• Journal of Korea Society of Industrial Information Systems
• /
• v.21 no.5
• /
• pp.1-10
• /
• 2016

In this paper, a car model recognition system using image processing and machine learning is proposed and it's performance is also evaluated. The system recognizes the front of car because the front of car is different for every car model and manufacturer, and difficult to remodel. The proposed method extracts HOG features from training data set, then builds classification model by the HOG features. If user takes photo of the front of car, then HOG features are extracted from the photo image and are used to determine the model of car based on the trained classification model. Experimental results show a high average recognition rate of 98%.

• Proceedings of the KIEE Conference
• /
• 2002.07b
• /
• pp.1283-1285
• /
• 2002

This study is developed vehicle information detector using loop and piezo sensors. This study would analyze the over all problems concerning our road conditions, environmental matters and unique features of our traffic matters; moreover, with these it would develope the hardware, software, car classification algorithm applied by artificial intelligence and traffic monitoring program which can be easily fixed. This can be divided into traffic detecting algorithm and car classification algorithm. Especially, we have developed the car classification algorithm used by C-means Fuzzy Clustering method.

• Journal of the Korea Convergence Society
• /
• v.9 no.6
• /
• pp.33-38
• /
• 2018

In this study, we try to detect new attacks for vehicle by learning only one class. We use Car-Hacking dataset, an intrusion detection dataset, which is used to evaluate classification performance. The dataset are created by logging CAN (Controller Area Network) traffic through OBD-II port from a real vehicle. The dataset have four attack types. One class classification is one of unsupervised learning methods that classifies attack class by learning only normal class. When using unsupervised learning, it difficult to achieve high efficiency because it does not use negative instances for learning. However, unsupervised learning has the advantage for classifying unlabeled data, which are new attacks. In this study, we use one class classifier to detect new attacks that are difficult to detect using signature-based rules on network intrusion detection system. The proposed method suggests a combination of parameters that detect all new attacks and show efficient classification performance for normal dataset.

• Journal of Auto-vehicle Safety Association
• /
• v.13 no.4
• /
• pp.53-59
• /
• 2021

According to the Korean Traffic Accident Analysis System (TAAS), more than 200,000 traffic accidents occur every year. Also, the statistics including auto insurance companies data show 1.3 million traffic accidents. In the case of TAAS, the types of traffic accidents are simply divided into four; frontal collision, side collision, rear collision, and rollover. However, more detailed information is needed to assess for advanced driver assist systems at intersections. For example, directional information is needed, such as whether the vehicle in the car accident way in a straight or a left turn, etc. This study intends to redefine the type of accident with the more clear driving direction and path by referring to the Negligence standards used in automobile insurance accidents. The standards largely divide five categories of car-to-car/motorcycle /pedestrian/cyclist, and highway, and the each category is classified into dozens of types by status of the traffic signal, conflict situations. In order to present more various accident types for auto insurance accidents, the standards are reclassified driving direction and path of vehicles from crash situations. In results, the car-to-car accidents are classified into 33 accident types, car-to-pedestrian accidents have 19 accident types, car-to-motorcycle accidents have 38 accident types, and car-to-cyclist accidents are derived into 26 types.

• Journal of Digital Convergence
• /
• v.15 no.3
• /
• pp.77-86
• /
• 2017

This study selected car-related mobile apps for app developers suffering from low revenue and classified car apps assisting users in driving or managing a car. A total of 697 car apps were classified into eight categories. Most apps are in four categories: car news & information (28%), locating service (23%), car rental service (15%), safe/efficient driving service (12%). The remaining categories are buying & selling, driver's communication, maintenance management, and expenses monitoring. Many apps are simple and too similar in their main functions. Only a few apps are designed to be more comprehensive and have functions in two or more categories. For the practicality of the categorization scheme, this study checked the inter-rater reliability in two tests and got 0.886 and 0.828. The result from this study suggests functions that are not implemented yet or need to be combined. Future research will focus on identifying promising car apps or designing multi-functional car apps.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.40 no.7
• /
• pp.582-589
• /
• 2012

This paper describes the algorithm recognizing car type from the image received from UAV and the recognition results between three types of car images. Using the NCC(Normalized Cross-Correlation) algorithm, geometric information is matched from template images. Template images are obtained from UAV and satellite map and indoor experiment is performed using satellite map. After verification of the possibility, experiment for verification of same car type recognition is performed using small UAV. In the experiment, same type cars are matched with 0.6 point similarity and truck with similar color distribution is not matched with template image of a sedan.

• Journal of Korean Society of Transportation
• /
• v.9 no.2
• /
• pp.5-26
• /
• 1991

Concepts of passenger car equivalent(PCE) based on various philosophies is reviewed. The headway method is redefined in two ways namely a macroscopic headway method and a microscopic headway method of which the concepts are confused in literature. As definition of the microscopic headway method is from intuitive approach so theoretical basis for this method is explained in connection with the macroscopic headway method. In addition new classification and evaluation of complicated existing theories was conducted to make the reader understand more easily.

• Proceedings of the Korean Society of Laser Processing Conference
• /
• 2003.05a
• /
• pp.39-42
• /
• 2003

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.6
• /
• pp.469-474
• /
• 2018

As urban population increases, research on urban environmental noise is getting more attention. In this study, we classify the abnormal noise occurring in traffic situation by using a deep learning algorithm which shows high performance in recent environmental noise classification studies. Specifically, we classify the four classes of tire skidding sounds, car crash sounds, car horn sounds, and normal sounds using convolutional neural networks. In addition, we add three environmental noises, including rain, wind and crowd noises, to our training data so that the classification model is more robust in real traffic situation with environmental noises. Experimental results show that the proposed traffic sound classification model achieves better performance than the existing algorithms, particularly under harsh conditions with environmental noises.

• Korean Journal of Computational Design and Engineering
• /
• v.13 no.3
• /
• pp.163-174
• /
• 2008

In this paper, for integrating of data on car parts we model information of parts that PDM system manages. Ontology of car parts applies existing ontology mapping research to integrate into car ontology. We propose a method for semantic integration of PLM object of MEMPHIS based on the integrated ontology. Through our method, we introduce C# ontology model to apply existing C# applications with ontology. We also classify ontology integration into three through examples and explain them. While semantically integrating PLM objects based on the integrated ontology, we explain the need for change of PLM object type and describe the process of change for PLM object type by examples.