Recently, the number of tunnels is increasing due to urbanization, and fire accidents in tunnels are likewise increasing. In particular, in a long tunnel of more than 1 km it is very difficult to track the exact location of a fire, accident vehicles, and the fire brigade, as well as whether a fire occurred. In this paper, we analyze various types of accidents that may occur in tunnel fires and propose detection, search, and rescue techniques to cope with them. For early detection of accidents, we propose various sensors using Internet of Things (IoT) technology and sensor networks to connect them. These sensors can detect not only a fire but also the position of the vehicle in which the fire is occurring in real time. We also propose a robotic system and operation technique that can be controlled by a fire fighter for more precise search operation. For rescue procedures, localization and tracking technology for fire fighters and robots is proposed. Finally, the efficiency of the proposed system was verified through actual performance tests, including simulations of actual placement and operation in tunnels. Through the construction of the equipment in an actual tunnel 1.9 km long, we show that the proposed system is good enough to cope with fire accidents, in terms of the delivery ratio of the collected data, fire recognition ratio, localization accuracy, and response delay.
Recently, radar technology has attracted attention for the realization of an intelligent transportation system (ITS) to monitor, track, and manage vehicle traffic on the roads as well as adaptive cruise control (ACC) and automatic emergency braking (AEB) for driving assistance of vehicles. However, when radar is installed on roads or in tunnels, the detection performance is significantly dependent on the deployment conditions and environment around the radar. In particular, in the case of tunnels, the detection accuracy for a moving vehicle drops sharply owing to the diffuse reflection of radio frequency (RF) signals. In this paper, we propose an optimal deployment condition based on height and tilt angle as well as a noise-filtering scheme for RF signals so that the performance of vehicle detection can be robust against external conditions on roads and in tunnels. To this end, first, we gather and analyze the misrecognition patterns of the radar by tracking a number of randomly selected vehicles on real roads. In order to overcome the limitations, we implement a novel road watch module (RWM) that is easily integrated into a conventional radar system such as Delphi ESR. The proposed system is able to perform real-time distributed data processing of the target vehicles by providing independent queues for each object of information that is incoming from the radar RF. Based on experiments with real roads and tunnels, the proposed scheme shows better performance than the conventional method with respect to the detection accuracy and delay time. The implemented system also provides a user-friendly interface to monitor and manage all traffic on roads and in tunnels. This will accelerate the popularization of future ITS services.
This paper proposes a vehicle light detection method using a hyperspectral camera instead of a Charge-Coupled Device (CCD) or Complementary metal-Oxide-Semiconductor (CMOS) camera for adaptive car headlamp control. To apply Intelligent Headlight Control (IHC), the vehicle headlights need to be detected. Headlights are comprised from a variety of lighting sources, such as Light Emitting Diodes (LEDs), High-intensity discharge (HID), and halogen lamps. In addition, rear lamps are made of LED and halogen lamp. This paper refers to the recent research in IHC. Some problems exist in the detection of headlights, such as erroneous detection of street lights or sign lights and the reflection plate of ego-car from CCD or CMOS images. To solve these problems, this study uses hyperspectral images because they have hundreds of bands and provide more information than a CCD or CMOS camera. Recent methods to detect headlights used the Spectral Angle Mapper (SAM), Spectral Correlation Mapper (SCM), and Euclidean Distance Mapper (EDM). The experimental results highlight the feasibility of the proposed method in three types of lights (LED, HID, and halogen).