Search

Accurate real-time traffic sensing is of key importance, especially in the urban environment to be able to optimise traffic flow by intelligent traffic systems (ITS). Often the high density of traffic sensors, needed to achieve an accurate real-time monitoring of important arterial roads, is difficult to implement due to technical contraints or because of high installation cost. Furthermore, existing traffic sensing technology uses sensors that are only able to measure traffic flow on a cross-section of the road where they are installed (typically on a junction), giving no information on the situation in between. An alternative "seamless" measuring technology, is to use floating car data, with Google Maps being the most prominant example. This technology allows to derive traffic information over wide road sections, however it is unable to deliver real- time information, and it relies on the “cooperation” of the data providers (the fleet owner or the mobile phone users). Fiber optic acoustic sensing (FOAS) is a new alternative technology that allows a seamless, real-time monitoring of the road traffic situation over large distances of up to 50 km using the existing telecom fiber optic cable infrastructure. In our previous work we presented an algorithm and results for traffic flow and average speed computation from FOAS raw data at a specific location along a highway and compared it to reference traffic data [1],[2]. In this paper we demonstrate the potential of the seamless nature of the technique by evaluating the traffic density over a length of 25 km of the monitored highway for different days and times of the day., Mobility, Knowledge and Innovation Hubs in Urban and Regional Development. Proceedings of REAL CORP 2022, 27th International Conference on Urban Development and Regional Planning in the Information Society, 465-470

This paper presents a new automatic vehicle detection method from very high resolution aerial images to measure traffic density. The proposed method starts by extracting road regions from image using road vector data. Then, the road image is divided into equal sections considering resolution of the images. Gradient vectors of the road image are computed from edge map of the corresponding image. Gradient vectors on the each boundary of the sections are divided where the gradient vectors significantly change their directions. Finally, number of vehicles in each section is carried out by calculating the standard deviation of the gradient vectors in each group and accepting the group as vehicle that has standard deviation above predefined threshold value. The proposed method was tested in four very high resolution aerial images acquired from Istanbul, Turkey which illustrate roads and vehicles with diverse characteristics. The results show the reliability of the proposed method in detecting vehicles by producing 86% overall F1 accuracy value., {"references":["P. Burlina, V. Parameswaran and R. Chellappa, \"Sensitivity Analysis\nand Learning Strategies for Context-Based Vehicle Detection\nAlgorithm,\" in DARPA IU Workshop, College Park, MD, Center for\nautomation Research, University of Maryland, 1997.","H. Moon, R. Chellappa and A. Rosenfeld, \"Performance Analysis of a\nSimple Vehicle Detection Algorithm,\" Image and Computer Vision,\nVol. 20, No. 1, pp. 1-13, 2002.","K. Kozempel and R. Reulke, \"Fast Vehicle Detection and Tracking in\nAerial Image Bursts\", in ISPRS City Models, Roads and Traffic\n(CMRT) , Paris, France, Vol. 38, No. 3/W4, pp. 175-180, 2009.","T. Zhao and R. Nevatia, \"Car Detection in Low Resolution Aerial\nImage\", Image and Vision Computing, Vol. 21, No. 8, pp. 693-703,\n2003.","S. Hinz, \"Detection of Vehicles and Vehicle Queues in High Resolution\nAerial Images\", Pjotogrammetrie-Fernerkundung-Geoinformation\n(PFG), Vol. 3, No. 4, pp. 201-213, 2004.","D. Lenhart, S. Hinz, J. Leitloff and U. Stilla, \"Automatic Traffic\nMonitoring Based On Aerial Image Sequences\", Pattern Recognition\nand Image Analysis, Vol. 18, No. 3, pp. 400-405, 2008.","J. Y. Choi and Y. K. Yang, \"Vehicle Detection from Aerial Images\nUsing Local Shape Information\", in Pacific Rim Symp. Advanced in\nImage and Video Technology (PSIVT) Berlin, Heidelberg, Germany,\nSpringer-Verlag, pp. 227-236, 2009.","A. C. Holt, E. Y. W. Seto, T. Rivard and G. Peng, \"Object-Based\nDetection and Classification of Vehicles from High-Resolution Aerial\nPhotography\", Photogrammetric Engineering and Remote Sensing (PE\n& RS), Vol. 75, No. 7, pp. 871-880, 2009.","M. Elmiktay and T. Stathaki, \"Car Detection in High-Resolution Urban\nScenes Using Multiple Image Descriptors\", in Proc. Of International\nConference on Pattern Recognition (ISPR), Stockholm, Sweden, pp.\n4299-4304, 2014.\n[10] X. Chen and Q. Meng, \"Vehicle Detection from UAVs by Using SIFT\nwith Implicit Shape Model\", in IEEE International Conference on\nSystems, Man, and Cybernetics, pp. 3139-3144, 2013.\n[11] T. Moranduzzo and F. Melgani, \"Detecting Cars in UAV Images with a\nCatalog-Based Approach,\" IEEE Transactions on Geoscience and\nRemote Sensing, Vol. 52, No. 10, pp. 6356-6367, 2014.\n[12] S. Tuermer, F. Kurz, P. Reinartz and U. Stilla, \"Airborne Vehicle\nDetection in Dense Urban Areas Using HoG Features and Disparity\nmaps,\" IEEE Journal of Selected Topics in Applied Earth Observation\nand Remote Sensing, Vol. 6, No. 6, pp. 2327-2337, 2013.\n[13] H. Grabner, T. T. Nguyen B. Gruber and H. Bischof, \"On-Line\nBoosting-Based Car Detection from Aerial Images\", ISPRS Journal of\nPhotogrammetry and Remote Sensing, Vol. 63, No. 3, pp. 382-396,\n2008.\n[14] S. Kluckner, G. Pacher, H. Garbner and H. Bischof, \"A 3D Teacher for\nCar Detection in Aerial Images\", Image and Vision Computing, in ICCV,\nRio de Janeiro, Brazil, 2007. [15] J. Leitloff, S. Hinz and U. Stilla, \"Vehicle Detection in Very High\nResolution Satellite Images of City Area,\" IEEE Transactions on\nGeoscience and Remote Sensing, Vol. 48, No. 7, pp. 2795-2806, 2010.\n[16] J. Canny, \"A Computational Approach to Edge Detection,\" in IEEE\nTransactions on Pattern Analysis and Machine Intelligence, Vol. 6, pp.\n697-698, 1986.\n[17] S. Ghaffarian and S. Ghaffarian, \"Automatic Building Detection Based\nOn Purposive FastICA (PFICA) Algorithm Using Monocular High\nResolution Google Earth Images,\" ISPRS Journal of Photogrammetry\nand Remote Sensing, Vol. 97, pp. 152-159, 2014.\n[18] S. Ghaffarian and S. Ghaffarian, \"Automatic Building Detection Based\nOn Supervised Classification Using High Resolution Google Earth\nImages,\" ISPRS Technical Commission III Symposium, Zurich,\nSwitzerland, pp. 101-106, 2014."]}