Centre de documentation
scientifique

Detection of roadway sign condition changes using multi-scale sign image matching (M-SIM) / Y.J. Tsai in Photogrammetric Engineering & Remote Sensing, PERS, vol 76 n° 4 (April 2010)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 76 n° 4 (April 2010) . - pp 391 - 405 Titre : Detection of roadway sign condition changes using multi-scale sign image matching (M-SIM) Type de document : Article/Communication Auteurs : Y.J. Tsai, Auteur ; Z. Hu, Auteur ; C. Alberti, Auteur Année de publication : 2010 Article en page(s) : pp 391 - 405 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] analyse d'image numérique [Termes IGN] appariement d'images [Termes IGN] classification [Termes IGN] détection de changement [Termes IGN] données multiéchelles [Termes IGN] image vidéo [Termes IGN] Louisiane (Etats-Unis) [Termes IGN] positionnement par GPS [Termes IGN] réseau routier [Termes IGN] signalisation routière Résumé : (Auteur) Roadway signs are important for safety, and transportation agencies need to identify sign condition changes to perform timely maintenance, including replacement. Currently, sign condition changes are inspected manually in the field, which is time consuming, costly, and some--times dangerous. This paper first proposes a novel algorithm to detect three condition changes: missing, tilted, and blocked signs, using GPS data and video log images. The algorithm consists of three steps: (a) Multi-Scale Sign Image Matching (m-sim), (b) Image feature analysis, and (c) Sign condition change detection and classification. The algorithm was tested using images with simulated sign condition changes and actual video images taken in Fiscal Year (fy) 2003 and 2005 by the Louisiana Department of Transportation and Development (ladotd). The tests demonstrate the algorithm is effective to detect three types of sign condition changes. Out of 34,000 actual video log images, the algorithm detected and classified 100 percent of the missing signs, 72.7 percent of the tilted signs, and 66.7 percent of the blocked signs, for an overall 74.3 percent detection rate. These results show that the algorithm is useful for developing an intelligent roadway sign condition change detection system. Copyright ASPRS Numéro de notice : A2010-120 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.76.4.391 En ligne : https://doi.org/10.14358/PERS.76.4.391 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=30316 [article]

Real-time speed limit sign recognition based on locally adaptive thresholding and depth-first-search / J. Wu in Photogrammetric Engineering & Remote Sensing, PERS, vol 71 n° 4 (April 2005)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 71 n° 4 (April 2005) . - pp 405 - 414 Titre : Real-time speed limit sign recognition based on locally adaptive thresholding and depth-first-search Type de document : Article/Communication Auteurs : J. Wu, Auteur ; Y.J. Tsai, Auteur Année de publication : 2005 Article en page(s) : pp 405 - 414 Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image [Termes IGN] acquisition de données [Termes IGN] image vidéo [Termes IGN] positionnement cinématique en temps réel [Termes IGN] reconnaissance de caractères [Termes IGN] réseau routier [Termes IGN] segmentation d'image [Termes IGN] seuillage d'image [Termes IGN] signalisation routière [Termes IGN] temps réel [Termes IGN] Visual C++ [Termes IGN] zone d'intérêt Résumé : (Auteur) Stop signs and speed limit signs (SLS) are the most popular and significant traffic signs on roadways. Unlike extracting stop signs with a distinct red color, extracting SLS in a realtime environment is much more challenging. This paper presents an algorithm for recognizing SLS from video imaging and extracting the numerical numbers of SLS to support real-time road inventory data collection operations. The algorithm consists of color segmentation based on locally adaptive thresholding extraction of regions of interest (ROI) using a depth-first-search algorithm, followed by speed limit sign detection and speed limit number extraction by means of optical character recognition and 2D correlation. The algorithm was implemented in Visual C++ language and tested on a non-Hyper-Threading Pentium IV PC with 3.06GHZ CPU using the images taken in the field with different image sizes. The average processing time for an image of 1200 X 800 pixels is about 125 ms. Experimental results from 1,401 video images show 0 percent false positives out of 1,278 images containing no SLS, and 3 percent false negatives out of 123 images containing SLS. Numéro de notice : A2005-159 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.71.4.405 En ligne : https://doi.org/10.14358/PERS.71.4.405 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=27297 [article]