{"id":30700,"key":"Commercial_softwares_to_extract_features_from_aerial_photos","title":"Commercial softwares to extract features from aerial photos","latest":{"id":1207460,"timestamp":"2025-11-27T22:17:34Z"},"content_model":"wikitext","license":{"url":"https://www.appropedia.org/Appropedia:Copyrights","title":"CC-BY-SA-4.0"},"source":"In digital photogrammetry,{{W|Photogrammetry}} features of objects are extracted using 3D information from image matching or DSM/DTM data, spectral, textural and other information sources. Pixel-based classification methods, either supervised or unsupervised, are mostly used for land-cover and man-made structure detections. For the classical methods e.g. minimum distance, parallelepiped and maximum likelihood, detailed information can be found in (Lillesand and Kiefer, 1994).\n\nIn general, the major difficulty in using aerial images is the complexity and variability of objects and their form, especially\nin suburban and densely populated urban regions (Weidner and Foerstner, 1995).\n\nObject identification and extraction from aerial photos can be done by scripting in C, command lines in the Image Analysis toolbox of Matlab or by using commercial softwares (with built functions): IMAGINE, ENVI, Feature Analyst extension of ArcFIS and GRASS.\n\nSupervised classification methods are preferable to\nunsupervised ones, because the target of the project is to detect well-defined standard target classes (airport buildings, bare\nground, grass, trees, roads, residential houses, shadows etc.),\npresent at airport sites. In [http://www.isprs.org/proceedings/XXXVIII/part3/b/pdf/131_XXXVIII-part3B.pdf Demir & Baltsavias 2010]\nthe training areas were selected\nmanually using AOI (Area of Interest) tools within the ERDAS\nImagine commercial software (Kloer, 1994). Among the\navailable image bands for classification (R, G and B from\ncolour images and NIR, R and G bands from CIR images), only\nthe bands from CIR images were used due to their better\nresolution and the presence of NIR channel (indispensable for\nvegetation detection). In addition, new synthetic bands were\ngenerated from the selected channels: a) 3 images from\nprincipal component analysis (PC1, PC2, PC3); b) one image\nfrom NDVI computation using the NIR-R channels and c) one\nsaturation image (S) obtained by converting the NIR-R-G\nchannels in the IHS (Intensity, Hue, Saturation) colour space.\nThe combination NIR-R-PC1-NDVI -S was selected for\nclassification using separability analysis. The maximum\nlikelihood classification method was used.\n\n== Reference ==\n\n* [http://www.isprs.org/proceedings/XXXVIII/part3/b/pdf/131_XXXVIII-part3B.pdf Demir & Baltsavias 2010]\n\n{{Page data\n| license = CC-BY-SA-3.0\n}}\n\n[[Category:Maps]]\n[[Category:Software]]"}