Title :
Classification of Urban Scenes from Geo-referenced Images in Urban Street-View Context
Author :
Iovan, Corina ; Picard, David ; Thome, Nicolas ; Cord, Matthieu
Author_Institution :
ECP-INRIA Saclay, Chtenay-Malabry, France
Abstract :
This paper addresses the challenging problem of scene classification in street-view georeferenced images of urban environments. More precisely, the goal of this task is semantic image classification, consisting in predicting in a given image, the presence or absence of a pre-defined class (e.g. shops, vegetation, etc.). The approach is based on the BOSSA representation, which enriches the Bag of Words (BoW) model, in conjunction with the Spatial Pyramid Matching scheme and kernel-based machine learning techniques. The proposed method handles problems that arise in large scale urban environments due to acquisition conditions (static and dynamic objects/pedestrians) combined with the continuous acquisition of data along the vehicle´s direction, the varying light conditions and strong occlusions (due to the presence of trees, traffic signs, cars, etc.) giving rise to high intra-class variability. Experiments were conducted on a large dataset of high resolution images collected from two main avenues from the 12th district in Paris and the approach shows promising results.
Keywords :
geographic information systems; image classification; learning (artificial intelligence); BOSSA representation; BoW model; bag of words model; intraclass variability; kernel-based machine learning; large scale urban environment; semantic image classification; spatial pyramid matching scheme; street-view georeferenced image; urban scene classification; urban street-view context; Dictionaries; Encoding; Feature extraction; Histograms; Kernel; Vegetation mapping; Visualization; kernel-based machine learning; semantic image classification; spatial pyramid matching; street-level images; visual words;
Conference_Titel :
Machine Learning and Applications (ICMLA), 2012 11th International Conference on
Conference_Location :
Boca Raton, FL
Print_ISBN :
978-1-4673-4651-1
DOI :
10.1109/ICMLA.2012.171