• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.416-421
• /
• 2021

In case of the road slips due to heavy snow and the temperature drops below 0 degrees, black ice which mainly occurs on the road, bridges for vehicles, and tunnel entrances, is not recognized by the driver's view because the image of the asphalt is transmitted through it. So cars' slip situation occurs, which leads to a big traffic accident and a large amount of loss of life and property. This study proposes a method to check the road condition using an infrared camera and to identify black ice through deep learning.

• Proceedings of the Korean Society of Disaster Information Conference
• /
• 2022.10a
• /
• pp.437-438
• /
• 2022

본 논문에서는 상변환 재료를 이용하여 동절기에 발생하는 도로포장의 블랙아이스 현상을 저감시키기 위한 기초연구를 진행하였다. 콘크리트 내에서 누출을 막기 위해 마이크로 캡슐화된 상변환 재료를 이용했으며 초기 개발단계인 만큼 페이스트 단계에서 실험을 진행 하였다. 실험은 초기 수화거동에서의 수화열 발생 경향을 분석하기 위해서 미소수화열 측정 실험과 반복되는 온도변화 환경에서의 상변환재료 캡슐의 안정성을 분석하기 위해 주사전자현미경을 이용하였다. 상변환 재료의 열에너지 저장 및 방출 능력은 시멘트의 미소수화열 발생에 상당한 변화를 주었으며, 온도변화에 의한 상변환 재료 캡슐계면의 이상은 발생하지 않는 것으로 나타났다.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.33-42
• /
• 2022

In this paper, we proposed a deep learning model based on multi-scale dilated convolution feature fusion for the segmentation of black ice region in road image to send black ice warning to drivers in real time. In the proposed multi-scale dilated convolution feature fusion network, different dilated ratio convolutions are connected in parallel in the encoder blocks, and different dilated ratios are used in different resolution feature maps, and multi-layer feature information are fused together. The multi-scale dilated convolution feature fusion improves the performance by diversifying and expending the receptive field of the network and by preserving detailed space information and enhancing the effectiveness of diated convolutions. The performance of the proposed network model was gradually improved with the increase of the number of dilated convolution branch. The mIoU value of the proposed method is 96.46%, which was higher than the existing networks such as U-Net, FCN, PSPNet, ENet, LinkNet. The parameter was 1,858K, which was 6 times smaller than the existing LinkNet model. From the experimental results of Jetson Nano, the FPS of the proposed method was 3.63, which can realize segmentation of black ice field in real time.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2022.04a
• /
• pp.170-171
• /
• 2022

In recent years, due to the occurrence of traffic accidents caused by black ice in winter, the number of personal injuries is increasing rapidly. Black ice is a phenomenon that occurs like a thin layer of ice on the road surface. Accordingly, many developments of heat-generating concrete are being developed to remove ice by increasing the temperature by supplying constant electricity to places where black ice is likely to occur. These heating elements are being developed by mixing a conductive material represented by carbon nanotubes with concrete. However, research up to now has been focused on efficient temperature rise and derivation of the optimum mixing ratio, and the evaluation of maintaining heat generation performance during continuous repetition is insufficient. Therefore, in this study, a heating test specimen was manufactured and 50V power was repeatedly supplied to evaluate the heating characteristics.

• Journal of the Korean Society of Industry Convergence
• /
• v.24 no.6_2
• /
• pp.873-882
• /
• 2021

In this study, a study was conducted to develop a system for predicting/responding to black ice occurring on roads in winter. Tests conditions were studied by making models of cement concrete pavement and asphalt concrete pavement. In order to freeze water on the manufactured model package, an tests was conducted at a temperature below zero using a freezer, and the freezing process was photographed using a thermal imaging camera. Black ice is generated when water is present on the road surface and the temperature is below freezing or the road surface temperature is below the dew point temperature. Under sub-zero conditions, the pavement, water, and ice were classified with a thermal imaging camera. As a result of the tests, it was possible to distinguish with a thermal imaging camera at a temperature below freezing in the same freezer due to the difference in the emissivity of the packaging, water, and ice. In the process of changing from water to ice during the tests, it was analyzed that ice and water were clearly distinguished by the thermal imaging camera due to the difference in emissivity and reflectance, so black ice could be predicted using the thermal imaging camera.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.524-526
• /
• 2021

As the invention of automobiles and construction of roads for vehicles began, the occurrence of traffic accidents began to increase. Accordingly, efforts were made to prevent traffic accidents by changing the road construction method and using signal systems such as traffic lights, but until now, numerous human and property damages have occurred every year due to traffic accidents caused by freezing of the road due to bad weather. In this paper, we propose a method of transmitting ice detection data detected using vehicle sensor data to vehicle navigation to reduce traffic accidents caused by road freezing.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.521-523
• /
• 2021

As the invention of automobiles and construction of roads for vehicles began, the occurrence of traffic accidents began to increase. Accordingly, efforts were made to prevent traffic accidents by changing the road construction method and using signal systems such as traffic lights, but even today, numerous human and property damages have occurred due to traffic accidents caused by freezing of the road due to bad weather. In this paper, in order to reduce traffic accidents due to road freezing, we propose a method of transferring the ice detection information obtained by deep learning of infrared wavelength data obtained using an infrared camera to the vehicle's navigation.

• Journal of Environmental Science International
• /
• v.26 no.7
• /
• pp.827-836
• /
• 2017

Freezing rain is a phenomenon when precipitation falls as a liquid rain drop, but freezes when it comes into contact with surfaces or objects. In this study, we investigated the predictability of freezing rain and its characteristics, which are strongly related with the occurrence of black ice using synoptic scale meteorological observation data. Two different cases occurred at 2012 were analyzed and in the presented cases, freezing rain often occurs in the low-level low pressure with the warm front. The warm front due to the lower cyclone make suitable environment in which snow falling from the upper layer can change into supercooled water. The $0^{\circ}C$ temperature line to generate supercooling water is located at an altitude of 850 hPa in the vertical temperature distribution. And the ground temperature remained below zero, as is commonly known as a condition for black ice formation. It is confirmed that the formation rate of freezing rain is higher when the thickness after 1000-850 hPa is 1290-1310 m and the thickness of 850-700 hPa layer is larger than 1540 m in both cases. It can also be used to predict and estimate the generation of freezing rain by detecting and analyzing bright bands in radar observation.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.10 no.1
• /
• pp.66-73
• /
• 2022

The purpose of this study was to develop self-heating concrete by utilizing the conduction resistance of concrete in order to reduce the risk of occurrence of black ice in the concrete pavement in winter and to prevent damage caused by freez-thawing effect. For this purpose, it was attempted to evaluate the strength and temperature exothermic characteristics using powder and liquid waste CNTs and a waste cathode agent as a conduction promotion. It was analyzed that liquid waste CNT had an effective dispersion degree in the mortar and a small decrease in strength occurred. In addition, DC 24 V was supplied by applying steel mesh, copper foil and copper wire to the mortar as electrodes, and the temperature change characteristics according to the mixing ratio of spent CNTs, anodes and carbon fibers were evaluated. In addition, by evaluating the temperature characteristics according to the electrode spacing from the selected optimal mixture, it was confirmed that it had sufficient heating characteristics up to an electrode spacing of 100 mm up to AC 50 V.

• Journal of the Korean Recycled Construction Resources Institute
• /
• v.8 no.4
• /
• pp.421-428
• /
• 2020

Recently, excellent thermal and electrical performance of cementitious composites by mixing nano materials are being studied. The purpose of this study is to research the heat generation and power consumption of rapid hardening nano-cementitious composites. The experiment was carried out after setting the rapid hardening cementitious material, curing day, and supply voltage as parameters. Rapid hardening nano-cementitious materials were classified into cement paste, mortar, and concrete The heat performance of all rapid hardening nano-cementitious composites in curing 1 day has increased over 10℃. The rapid hardening nano-cementitious composites can exhibit heat performance within 1 day. The heat performance of the rapid hardening nano-cementitious composites is maintained after 28 days.