Intelligent Road Signs Research Articles

Integration of Image-Based Fog Detection with Autonomous Decision System for Intelligent Road Sign

Abstract The paper presents the description of the decision system implemented for Intelligent Road Signs. It focuses on the implementation of the novel air transparency analysis system and its integration with the rule system and the speed control infrastructure. Moreover, there are presented issues of making decisions about the content displayed in the case of autonomous and cooperating signs. To reflect more closely on real-life situations, it is assumed that the content presented by the IRS changes dynamically, depending on the road traffic and weather parameters. The IRS system operation was presented using fog detection as an example.

Comparing traffic intensity estimates employing passive acoustic Radar and microwave Doppler Radar sensor

The purpose of our applied research project is to develop an autonomous road sign with built-in radar devices of our design. In this paper, we show that it is possible to calibrate the acoustic vector sensor in such a way that it can be used to measure traffic volume and to count the vehicles involved in the traffic through the analysis of the noise emitted by them. Signals obtained from a Doppler radar are used as a reference source. Although the acoustical vector sensor (AVS), being the embodiment of acoustic radar, has a lower accuracy than Doppler radar in vehicle counting and it is not able to measure the vehicle speed with the same precision, it has some advantages over the Doppler sensor. Namely, it does not emit any signals, it is not susceptible to electromagnetic interferences and it allows for further analysis of audio signals, such as assessment of the road surface state (e.g., wet/dry). The acoustic radar we developed is a new proposition of the acoustic method for road traffic monitoring. In addition, our research allowed a comparison of the efficiency of both methods, i.e., microwave and acoustic ones. [Project No. POIR.04.01.04-0089/16 entitled: “INZNAK – Intelligent road signs with V2X interface for adaptive traffic controlling” is financed by the Polish National Centre for Research and Development from under the EU Operational Programme Innovative Economy.]

An acoustic-based road traffic monitoring method

The paper shows an acoustic-based road traffic monitoring method and its capabilities of utilizing it in the road system management. First, the acoustical signal acquisition process is shown. The database development and the methodology of audio signals preprocessing are described. Next, selecting appropriate parameters used to detect the presence of the vehicle on the road is discussed. The detected events are subjected first to the statistical evaluation related to intervals between vehicles (safe distance criterion), and then classification is carried out. Several methods for classifying vehicle types based on their sound are employed, using four classes of vehicles. Additionally, the method of assessing traffic safety is presented with regard to the vehicle tire adhesion with the road surface—by acoustic detection of the presence of water and snow on the roadway. The effectiveness of 0,968 has been obtained for the type of a vehicle, whereas 93.3% for detecting humidity on the road surface. [The project is financed by the Polish National Centre for Research and Development (NCBR) from the European Regional Development Fund under the Operational Programme Innovative Economy No. POIR.04.01.04-00-0089/16 entitled: INZNAK – “Intelligent road signs for adaptive traffic control, communicating in V2X technology.”]

System for monitoring road slippery based on CCTV cameras and convolutional neural networks

The slipperiness of the surface is essential for road safety. The growing number of CCTV cameras opens the possibility of using them to automatically detect the slippery surface and inform road users about it. This paper presents a system of developed intelligent road signs, including a detector based on convolutional neural networks (CNNs) and the transfer-learning method employed to the processing of images acquired with video cameras. Based on photos taken in different light conditions by CCTV cameras located at the roadsides in Poland, four network topologies have been trained and tested: Resnet50 v2, Resnet152 v2, Vgg19, and Densenet201. The last-mentioned network has proved to give the best result with 98.34% accuracy of classification dry, wet, and snowy roads.

To Perform Road Signs Recognition for Autonomous Vehicles Using Cascaded Deep Learning Pipeline

Autonomous vehicle is a vehicle that can guide itself without human conduction. It is capable of sensing its environment and moving with little or no human input. This kind of vehicle has become a concrete reality and may pave the way for future systems where computers take over the art of driving. Advanced artificial intelligence control systems interpret sensory information to identify appropriate navigation paths, as well as obstacles and relevant road signs. In this paper, we introduce an intelligent road signs classifier to help autonomous vehicles to recognize and understand road signs. The road signs classifier based on an artificial intelligence technique. In particular, a deep learning model is used, Convolutional Neural Networks (CNN). CNN is a widely used Deep Learning model to solve pattern recognition problems like image classification and object detection. CNN has successfully used to solve computer vision problems because of its methodology in processing images that are similar to the human brain decision making. The evaluation of the proposed pipeline was trained and tested using two different datasets. The proposed CNNs achieved high performance in road sign classification with a validation accuracy of 99.8% and a testing accuracy of 99.6%. The proposed method can be easily implemented for real time application.

Counting and tracking vehicles using acoustic vector sensors

A method is presented for counting vehicles and for determining their movement direction by means of acoustic vector sensor application. The assumptions of the method employing spatial distribution of sound intensity determined with the help of an integrated 3D intensity probe are discussed. The intensity probe developed by the authors was used for the experiments. The mode of operation of the algorithm is presented in conjunction with noise characteristics produced by moving vehicles. The optimization of the algorithm is based on measurements of intensity of sound emitted by the vehicle under controlled conditions. A test setup was built for this purpose with the use of measuring devices installed along a road with varying traffic flow. Reference data on the number of vehicles and traffic directions were prepared employing a recorded video and a reference traffic analyzer operating in lidar technology. It is shown that the developed acoustic method may contribute to an increase of effectiveness of commonly used vehicle counting systems employing inductive loops or Doppler radars. [Project financed by the by the Polish National Centre for Research and Development (NCBR) from the European Regional Development Fund under the Operational Programme Innovative Economy No. POIR.04.01.04-00-0089/16 “INZNAK—Intelligent road signs...”.]A method is presented for counting vehicles and for determining their movement direction by means of acoustic vector sensor application. The assumptions of the method employing spatial distribution of sound intensity determined with the help of an integrated 3D intensity probe are discussed. The intensity probe developed by the authors was used for the experiments. The mode of operation of the algorithm is presented in conjunction with noise characteristics produced by moving vehicles. The optimization of the algorithm is based on measurements of intensity of sound emitted by the vehicle under controlled conditions. A test setup was built for this purpose with the use of measuring devices installed along a road with varying traffic flow. Reference data on the number of vehicles and traffic directions were prepared employing a recorded video and a reference traffic analyzer operating in lidar technology. It is shown that the developed acoustic method may contribute to an increase of effectiveness of common...

A Video Camera Road Sign System of the Early Warning from Collision with the Wild Animals

Abstract This paper proposes a camera road sign system of the early warning, which can help to avoid from vehicle collision with the wild animals. The system consists of camera modules placed down the particularly chosen route and the intelligent road signs. The camera module consists of the camera device and the computing unit. The video stream is captured from video camera using computing unit. Then the algorithms of object detection are deployed. Afterwards, the machine learning algorithms will be used to classify the moving objects. If the moving object is classified as animal and this animal can be dangerous for safety of the vehicle, warning will be displayed on the intelligent road sings.

Generalized Sign Recognition Based on the Gaussian Statistical Color Model for Intelligent Road Sign Inventory

The rapid progress in sensor and information technologies makes it possible to develop intelligent road sign inventory (IRSI) by automatically collecting and processing vehicle-borne multisensor data. This study developed a framework for IRSI with a vehicle-borne multisensor system. The IRSI framework consists of four modules: ( a) sensor vehicle design for multisensor data collection, ( b) sign recognition, ( c) sign data integration and attribute computation, and ( d) IRSI-based applications. Within the framework, sign recognition is the key to IRSI. The study proposed a novel Gaussian statistical color model (G-SCM) for image segmentation and color feature extraction. The G-SCM has good capability to model actual sign color distributions and has fast implementation for G-SCM-based color segmentation. With color features integrated into sign texture, shape, location, and other features, a novel generalized recognition algorithm is proposed for sign inventory. The proposed G-SCM model and the sign recognition method were tested with actual video log images. The experimental results show that the G-SCM can model actual sign colors and segment sign images accurately. The fast implementation can greatly enhance computation speed compared with existing color analysis methods. The developed sign recognition algorithm was tested to recognize all speed limit signs from video log images collected along a 140-km highway segment. The algorithm recognized 30 of 31 signs and achieved a 96.8% site recognition rate. Of the 136 images containing speed limit signs, the algorithm had only eight false positives. The results demonstrate that the proposed sign recognition method is promising for the development of IRSI.

Intelligent Road Sign Inventory (IRSI) with Image Recognition and Attribute Computation from Video Log

Abstract: Road signs are infrastructure assets crucial for transportation safety. Transportation agencies need to inventory signs regularly, e.g., annually or every 2 years, for asset management. This article proposes a framework of intelligent road sign inventory (IRSI) to enhance sign data collection, compared with existing manual or semiautomatic methods. The two key technical problems of sign recognition and attribute computation are addressed for IRSI. Because sign colors are important for both image segmentation and color features analysis, a novel kernel‐density‐estimation‐based statistical color model (KDE‐SCM) is proposed to describe the widespread and multimodel sign color distributions. The KDE‐SCM is embedded into a sign recognition algorithm by integrating sign shapes, textures, and other features. Furthermore, sign attributes, including the distance, local position, global positioning system (GPS) position, height, tilt angle, etc., are computed from image and GPS data. The proposed algorithms have been tested with the actual video log images collected on 35 km road segments in Kyoto City, Japan. For the KDE‐SCM color models, the false positive rates for the red and blue colors are 0.9% and 0.7%, while the false negative rates are 0.8% and 0.5%, respectively. The sign recognition algorithm was configured and applied to recognize three regulatory sign types of stop, no‐crossing, and no‐parking signs with 92.6% precision and 91.9% recall on average. The sign attribute computation model was also tested with the actual video log. The computation errors for distance, position, tilt angle, and height, etc., increase with the measurement distances. The sign‐to‐camera distances have the computation errors of less than 22 cm if the measurement distances are less than 15 m. The proposed algorithms demonstrate good potential for developing IRSI to enhance road asset data collection and processing.