Centre dedocumentation
scientifique

Using geometric constraints to improve performance of image classifiers for automatic segmentation of traffic signs / Roholah Yazdan in Geomatica [en ligne], vol 75 n° 1 (Mars 2021)  

Public [article]  inGeomatica [en ligne] > vol 75 n° 1 (Mars 2021) . - pp 28 - 50 Titre : Using geometric constraints to improve performance of image classifiers for automatic segmentation of traffic signs Type de document : Article/Communication Auteurs : Roholah Yazdan, Auteur ; Masood Varshosaz, Auteur ; Saied Pirasteh, Auteur ; Fabio Remondino, Auteur Année de publication : 2021 Article en page(s) : pp 28 - 50 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] classification par séparateurs à vaste marge [Termes IGN] contrainte géométrique [Termes IGN] espace colorimétrique [Termes IGN] programmation par contraintes [Termes IGN] signalisation routière Résumé : (auteur) Automatic detection and recognition of traffic signs from images is an important topic in many applications. At first, we segmented the images using a classification algorithm to delineate the areas where the signs are more likely to be found. In this regard, shadows, objects having similar colours, and extreme illumination changes can significantly affect the segmentation results. We propose a new shape-based algorithm to improve the accuracy of the segmentation. The algorithm works by incorporating the sign geometry to filter out the wrong pixels from the classification results. We performed several tests to compare the performance of our algorithm against those obtained by popular techniques such as Support Vector Machine (SVM), K-Means, and K-Nearest Neighbours. In these tests, to overcome the unwanted illumination effects, the images are transformed into colour spaces Hue, Saturation, and Intensity, YUV, normalized red green blue, and Gaussian. Among the traditional techniques used in this study, the best results were obtained with SVM applied to the images transformed into the Gaussian colour space. The comparison results also suggested that by adding the geometric constraints proposed in this study, the quality of sign image segmentation is improved by 10%–25%. We also comparted the SVM classifier enhanced by incorporating the geometry of signs with a U-Shaped deep learning algorithm. Results suggested the performance of both techniques is very close. Perhaps the deep learning results could be improved if a more comprehensive data set is provided. Numéro de notice : A2021-608 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1139/geomat-2020-0010 En ligne : https://doi.org/10.1139/geomat-2020-0010 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=98322 [article]

Improving traffic sign recognition results in urban areas by overcoming the impact of scale and rotation / Roholah Yazdan in ISPRS Journal of photogrammetry and remote sensing, vol 171 (January 2021)  

Public [article]  inISPRS Journal of photogrammetry and remote sensing > vol 171 (January 2021) . - pp 18 - 35 Titre : Improving traffic sign recognition results in urban areas by overcoming the impact of scale and rotation Type de document : Article/Communication Auteurs : Roholah Yazdan, Auteur ; Masood Varshosaz, Auteur Année de publication : 2021 Article en page(s) : pp 18 - 35 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage profond [Termes IGN] base de données d'images [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] classification par séparateurs à vaste marge [Termes IGN] corrélation à l'aide de traits caractéristiques [Termes IGN] corrélation croisée normalisée [Termes IGN] couple stéréoscopique [Termes IGN] détection automatique [Termes IGN] modèle stéréoscopique [Termes IGN] reconnaissance d'objets [Termes IGN] segmentation d'image [Termes IGN] SIFT (algorithme) [Termes IGN] signalisation routière [Termes IGN] SURF (algorithme) [Termes IGN] Téhéran [Termes IGN] transformation de Hough [Termes IGN] zone urbaine Résumé : (auteur) Automatic detection and recognition of traffic signs have many applications. However, some problems can affect the accuracy of the existing algorithms, such as changes in environmental light conditions, shadows, the presence of objects of the same colour, significant changes in scale and rotation, as well as obstacles in front of the traffic signs. To overcome these difficulties, a reference image database is usually used that includes different modes of appearing the traffic signs in the images. In order to overcome the effects of scale and rotation, in this paper a new method is presented in which only one reference image is needed for each sign to recognise the traffic sign in an image. In the proposed method, imaging is done in stereo. Using the captured image pair, a virtual image is generated which is then used to recognise the sign. As a result, the recognition is carried out with a minimum number of reference images. Experiments show that the proposed algorithm significantly improves recognition results. The traffic signs are recognised with 93.1% accuracy that enjoys a 4.9% improvement over traditional methods. Numéro de notice : A2021-010 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1016/j.isprsjprs.2020.10.003 Date de publication en ligne : 06/11/2020 En ligne : https://doi.org/10.1016/j.isprsjprs.2020.10.003 Format de la ressource électronique : url article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=96304 [article]

Réservation

Réserver ce document

Exemplaires (3)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
081-2021011	SL	Revue	Centre de documentation	Revues en salle	Disponible
081-2021013	DEP-RECP	Revue	LaSTIG	Dépôt en unité	Exclu du prêt
081-2021012	DEP-RECF	Revue	Nancy	Dépôt en unité	Exclu du prêt

Optimal placement of a terrestrial laser scanner with an emphasis on reducing occlusions / Morteza Heidari Mozaffar in Photogrammetric record, vol 31 n° 156 (December 2016 - February 2017)  

Public [article]  inPhotogrammetric record > vol 31 n° 156 (December 2016 - February 2017) . - pp 374 – 393 Titre : Optimal placement of a terrestrial laser scanner with an emphasis on reducing occlusions Type de document : Article/Communication Auteurs : Morteza Heidari Mozaffar, Auteur ; Masood Varshosaz, Auteur Année de publication : 2016 Article en page(s) : pp 374 – 393 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Acquisition d'image(s) et de donnée(s) [Termes IGN] angle de visée [Termes IGN] occultation du signal [Termes IGN] optimisation spatiale [Termes IGN] point de visibilité [Termes IGN] télémétrie laser terrestre Résumé : (auteur) In this paper, a new automated algorithm is proposed that finds the optimum locations of a terrestrial laser scanner (TLS), ensuring completeness of data and minimising the number of scanning locations. The process starts with an initial scan and placing a 3D grid of candidate stations over the entire scan area. A global visibility analysis is then performed to identify the next best view (NBV) location. The TLS is placed on this selected point and a new scan is recorded. Having updated the initial scan with the resulting point cloud, the model is checked for completeness and density. The process is repeated until full coverage of the scan area is achieved by determining the best global arrangement with the minimum number of stations. Experiments show that the algorithm is able to automatically determine the station positions and provide a coverage of 99·5% for simulated data and 91% for real data. Numéro de notice : A2016--002 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1111/phor.12162 Date de publication en ligne : 15/12/2016 En ligne : https://doi.org/10.1111/phor.12162 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=83820 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
106-2016041	SL	Revue	Centre de documentation	Revues en salle	Disponible

Development of a new stereo-panorama system based on off-the-shelf stereo cameras / Amir Shahrokh Amini in Photogrammetric record, vol 29 n° 146 (June - August 2014)  

Public [article]  inPhotogrammetric record > vol 29 n° 146 (June - August 2014) . - pp 206 - 223 Titre : Development of a new stereo-panorama system based on off-the-shelf stereo cameras Type de document : Article/Communication Auteurs : Amir Shahrokh Amini, Auteur ; Masood Varshosaz, Auteur ; Mohammad Saadatseresht, Auteur Année de publication : 2014 Article en page(s) : pp 206 - 223 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Photogrammétrie numérique [Termes IGN] image panoramique [Termes IGN] mesure géométrique [Termes IGN] modèle stéréoscopique [Termes IGN] modélisation 3D Résumé : (Auteur) There exist a number of techniques which are used to produce stereo panoramas, each of which has its own advantages and disadvantages. In this paper, a new stereo-panorama generating system is presented which uses off-the-shelf stereo cameras. In addition to stereoviewing, the system offers the possibility of making geometric measurements. The benefits of such a system are: (a) overcoming the effects of moving objects in the images; (b) increased automation of matching and stitching; (c) simple geometric information extraction; and (d) equal resolution over the whole stereo panorama. The structure of the system is described, including system implementation, calibration, stereo-panorama generation, georeferencing, co-registration of multiple panoramas and 3D measurements. In addition, an experiment to evaluate the capability of the system in modelling 3D environments is reported. Evaluation results show that the system can produce a suitable 3D panoramic environment for the simultaneous 3D visualisation and measurement of objects, especially in urban areas. Numéro de notice : A2014-281 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1111/phor.12059 Date de publication en ligne : 15/06/2014 En ligne : https://doi.org/10.1111/phor.12059 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=33184 [article]

Development of a new laser triangulation system based on an optical frame of reference / A. Abzal in Photogrammetric record, vol 26 n° 135 (September - November 2011)  

Public [article]  inPhotogrammetric record > vol 26 n° 135 (September - November 2011) . - pp 293 - 306 Titre : Development of a new laser triangulation system based on an optical frame of reference Type de document : Article/Communication Auteurs : A. Abzal, Auteur ; Masood Varshosaz, Auteur ; M. Saadatseresht, Auteur Année de publication : 2011 Article en page(s) : pp 293 - 306 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Lasergrammétrie [Termes IGN] balayage laser [Termes IGN] caméra numérique [Termes IGN] lasergrammétrie [Termes IGN] photographie terrestre [Termes IGN] précision géométrique (imagerie) [Termes IGN] triangulation locale Résumé : (Auteur) In this paper a new triangulation-based laser scanner is presented which has a simple, yet strong, flexible and low-cost structure. A digital camera and three laser line projectors are the main components of the system. One of the laser projectors is positioned vertically, while the other two are horizontal. The former scans the object, whereas the latter two establish an optical frame which is used, in part, to define the plane containing the vertical laser projector at each step of scanning. At each step, an image is taken which includes the object along with the projected laser lines. By intersecting the vertical and horizontal lines a couple of points are formed which, along with the calibration information of the system, enable the extraction of the object coordinates. Results of the tests carried out show that by using an optical frame of this nature, the process of scanning is greatly facilitated. That is, the scanner can easily be used to scan objects of different size and dimensions. Within the current configuration, the system enables measurements with an accuracy of 1/1600. Also, as the system has a rigorous basis, its accuracy can be increased if improved hardware is provided. Numéro de notice : A2011-385 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1111/j.1477-9730.2011.00646.x Date de publication en ligne : 14/09/2011 En ligne : https://doi.org/10.1111/j.1477-9730.2011.00646.x Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=31164 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
106-2011031	RAB	Revue	Centre de documentation	En réserve 3L	Disponible

Visibility prediction based on artificial neural networks used in automatic network design / M. Saadatseresht in Photogrammetric record, vol 22 n° 120 (December 2007 - February 2008)  

Permalink