Generic rule-sets for automated detection of urban tree species from very high-resolution satellite data

Abstract

The sustainable management and monitoring of urban forests is an important activity in the urbanized world, and operational approaches require information about the status of urban trees to determine the best strategy. One limitation in urban forest studies is the detection and discrimination of tree species using limited training data. Thus, this study focuses on developing generic rule sets from high-resolution WorldView-2 imagery in conjunction with spectral, spatial, colour and textural information for automated urban tree species detection. The object-based image analysis and its combination with statistical analysis of object features is utilized for this purpose. Results of attribute selection indicated that from 55 attributes, only 26 were useful to discriminate urban tree species, namely Messua ferrea L., Samanea saman and Casuarina sumatrana. Finally, the high overall accuracy, approximately 86.87% with kappa of 0.75 confirmed the transferability of the generic model.

Impact of Urban Vegetation in Sustainable Smart Cities Development: A Comprehensive Appraisal

“…On the other hand, a review conducted on tree species classification without specifically focusing on urban areas. They found that many studies employed data-driven approaches but lacked a clear target in terms of anticipated applications or accuracy [13].This literature review aims to provide a comprehensive overview of state-of-the-art methods and approaches used for mapping different types of urban green spaces using high-resolution satellite images data. The focus is on spatially and thematically detailed mapping of urban vegetation and the various mapping methodologies employed.To compile the papers for this review study, a limited number of search queries were used, and the "snowballing" approach was applied to expand the database.…”

Suivi de la dynamique de deux parcs à Faidherbia albida du Nord-Cameroun, par analyse d’images Google Earth

“…Pour ces raisons, nous avons opté dans cette étude pour une extraction des couverts arborés par une photo-interprétation semi-automatique grâce au logiciel ENVI EX. Cette approche, largement utilisée (Mo et al, 2012 ;Shojanoori et al, 2018), est basée sur une analyse orientée-objet (Blaschke et Hay, 2001). Dans cette analyse, l'extraction des objets (couverts arborés dans notre cas) prend en considération non seulement les attributs spectraux et de couleur, mais aussi les attributs spatiaux et de texture (Benz et al, 2004 ;Rittl et al, 2013 ;Shojanoori et al, 2018).…”

“…Cette approche, largement utilisée (Mo et al, 2012 ;Shojanoori et al, 2018), est basée sur une analyse orientée-objet (Blaschke et Hay, 2001). Dans cette analyse, l'extraction des objets (couverts arborés dans notre cas) prend en considération non seulement les attributs spectraux et de couleur, mais aussi les attributs spatiaux et de texture (Benz et al, 2004 ;Rittl et al, 2013 ;Shojanoori et al, 2018). Pour ce faire, d'abord, après une segmentation de chaque image en petites régions de pixels, les régions sont fusionnées selon un seuil défini et jugé optimal par le cartographe pour bien délimiter les objets recherchés (ITT Visual Information Solutions, 2009).…”

Remote sensing of urban green spaces: A review

“…Based on recent advances such as high spatial resolution imagery and free data access policies, remote sensing is providing a valuable set of tools which are able to minimize the need for field survey, even in highly heterogeneous and complex urban settings. For instance, remote sensing has proven to be effective for mapping street trees (Parmehr et al, 2016), detecting species within UGSs (Shojanoori et al, 2018), mapping invasive shrubs in UGSs (Chance et al, 2016) and assessing vegetation health within UGSs (Nasi et al, 2018). Furthermore, current remote sensing programs such as Copernicus (Harris and Baumann, 2015) and Landsat (Zhu et al, 2019) not only provide historical time-series data but also facilitate access to recently acquired data.…”