• Korean Journal of Artificial Intelligence
• /
• 제11권4호
• /
• pp.1-8
• /
• 2023

In this paper, we propose a black ice detection platform framework using Convolutional Neural Networks (CNNs). To overcome black ice problem, we introduce a real-time based early warning platform using CNN-based architecture, and furthermore, in order to enhance the accuracy of black ice detection, we apply a multi-scale dilation convolution feature fusion (MsDC-FF) technique. Then, we establish a specialized experimental platform by using a comprehensive dataset of thermal road black ice images for a training and evaluation purpose. Experimental results of a real-time black ice detection platform show the better performance of our proposed network model compared to conventional image segmentation models. Our proposed platform have achieved real-time segmentation of road black ice areas by deploying a road black ice area segmentation network on the edge device Jetson Nano devices. This approach in parallel using multi-scale dilated convolutions with different dilation rates had faster segmentation speeds due to its smaller model parameters. The proposed MsCD-FF Net(2) model had the fastest segmentation speed at 5.53 frame per second (FPS). Thereby encouraging safe driving for motorists and providing decision support for road surface management in the road traffic monitoring department.

• Journal of Korean Tunnelling and Underground Space Association
• /
• 제20권3호
• /
• pp.561-573
• /
• 2018

In order to design the ventilation in the road tunnel, it is necessary to know the ratio of average annual daily traffic by vehicle type. In general, the road tunnels are onedirectional tunnel, so the traffic of each vehicle type does not change along the tunnel. On the other hand, in the case of network road tunnels, since the connections in the tunnels are complex, the traffic of vehicle-type varies depending on the network composition of tunnels. In the studying the easy method for calculating the ratio of vehicle type for the network road tunnel are proposed with using incidence matrix.

• Journal of the Korean Society of Systems Engineering
• /
• 제1권2호
• /
• pp.26-33
• /
• 2005

In this paper, an algorithm is presented to recognize road based on unpaved test courses image. The road images obtained by a video camera undergoes a pre-processing that includes filtering, gray level slicing, masking and identification of unpaved test courses. After this pre-processing, a part of image is grouped into 27 sub-windows and fed into a three-layer feed-forward neural network. The neural network is trained to indicate the road direction. The proposed algorithm has been tested with the images different from the training images, and demonstrated its efficacy for recognizing unpaved road. Based on the test results, it can be said that the algorithm successfully combines the traditional image processing and the neural network principles towards a simpler and more efficient driver warning or assistance system.

• Korean Journal of Artificial Intelligence
• /
• 제11권3호
• /
• pp.17-22
• /
• 2023

In this paper, we propose a black ice detection system using Convolutional Neural Networks (CNNs). Black ice poses a serious threat to road safety, particularly during winter conditions. To overcome this problem, we introduce a CNN-based architecture for real-time black ice detection with an encoder-decoder network, specifically designed for real-time black ice detection using thermal images. To train the network, we establish a specialized experimental platform to capture thermal images of various black ice formations on diverse road surfaces, including cement and asphalt. This enables us to curate a comprehensive dataset of thermal road black ice images for a training and evaluation purpose. Additionally, in order to enhance the accuracy of black ice detection, we propose a multi-scale dilation convolution feature fusion (MsDC-FF) technique. This proposed technique dynamically adjusts the dilation ratios based on the input image's resolution, improving the network's ability to capture fine-grained details. Experimental results demonstrate the superior performance of our proposed network model compared to conventional image segmentation models. Our model achieved an mIoU of 95.93%, while LinkNet achieved an mIoU of 95.39%. Therefore, it is concluded that the proposed model in this paper could offer a promising solution for real-time black ice detection, thereby enhancing road safety during winter conditions.

• Proceedings of the KIEE Conference
• /
• 대한전기학회 1991년도 하계학술대회 논문집
• /
• pp.837-840
• /
• 1991

This paper describes an implementation of road network extraction algorithms for satellite images. We propose a new road network extraction algorithm which uses magnitude and direction information of edges. The results of applying the proposed algorithm to satellite images are presented and compared with those of other algorithms.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 한국정밀공학회 1995년도 추계학술대회 논문집
• /
• pp.341-343
• /
• 1995

In this paper, an algorithm is presented to recogniz the driving direction of a vehicle and obstacles in front of it based on highway road image. The algorithm employs a neural network with 27 sub sets obtained from the road image as its input. The outputs include the direction of the vehicle movement and presence or absence of obstacles. The road image, obtained by a video camera, was digitized and processed by a personal computer equipped with an image processing board.

• Journal of Korean Society of Rural Planning
• /
• 제11권1호
• /
• pp.1-8
• /
• 2005

The purpose of this study is to develop a simulation model of rural road path for infrastructure of green-tourism and public service in rural areas. This study makes an objective function for moving cost minimization considering car travel time according to road characteristics, which can route the optimal shortest road paths between the center places and all rear villages, based on GIS coverages of road-village network for connecting between center places and rural villages as input data of the model. In order to verify the model algorithm, a homogeneous hexagonal network, assuming distribution of villages with same population density and equal distance between neighborhood villages on a level plane area, was tested to simulate the optimal paths between the selected center nodes and the other rear nodes, so that the test showed reasonable shortest paths and road intensity defined in this study. The model was also applied to the actual rural area, Ucheon-myun, which is located on Hoengsung-gun, Kangwon-do, with 72 rural villages, a center village (Uhang, 1st center place) in the area, a county conte. (Hoengsung-eup, 2nd center place), and a city (Wonju, 3rd center place), as upper settlement system. The three kinds of conte. place, Uhang, Hoengsung-eup, and Wonju, were considered as center places of three scenarios to simulate the optimal shortest paths between the centers and rural villages, respectively. The simulation results on the road-village network with road information about pavement and width of road show that several spans having high intensity of road are more important that the others, while some road spans have low intensity of road.

• Journal of information and communication convergence engineering
• /
• 제20권4호
• /
• pp.273-279
• /
• 2022

Owing to the need to establish a cooperative-intelligent transport system (C-ITS) environment in the transportation sector locally and abroad, various research and development efforts such as high-tech road infrastructure, connection technology between road components, and traffic information systems are currently underway. However, the current central control center-oriented information collection and provision service structure and the insufficient road infrastructure limit the realization of the C-ITS, which requires a diversity of traffic information, real-time data, advanced traffic safety management, and transportation convenience services. In this study, a network construction method based on the existing received signal strength indicator (RSSI) selected as a comparison target, and the experimental target and the proposed intelligent edge network compared and analyzed. The result of the analysis showed that the data transmission rate in the intelligent edge network was 97.48%, the data transmission time was 215 ms, and the recovery time of network failure was 49,983 ms.

• Transactions of the Korean Society of Automotive Engineers
• /
• 제10권5호
• /
• pp.147-159
• /
• 2002

This paper proposes a method to recognize the noisiness of road images connected with the extraction of lane-related information in order to prevent the usage of erroneous information. The proposed method uses a fuzzy neural network(FNN) with the back-Propagation loaming algorithm. The U decides road images good or bad with respect to visibility of lane marks on road images. Most input parameters to the FNN are extracted from an edge distribution function(EDF), a function of edge histogram constructed by edge phase and norm. The shape of the EDF is deeply correlated to the visibility of lane marks of road image. Experimental results obtained by simulations with real images taken by various lighting and weather conditions show that the proposed method was quite successful, providing decision-making of noisiness with about 99%.

• Spatial Information Research
• /
• 제3권1호
• /
• pp.55-63
• /
• 1995

The database design, which logically sets the base structure and orga¬nization of the database, is performed by considering the users requirement, the relations between various data, and the relations between data and application field.The road network data must be created to have geometrical topological structure, because various data elements are needed to recognize the state of each section and to relate between data element. In this study, we propose a method of creating the road network database for an integrated road management system.