No abstract

Temporal and spatial resolution requirements for extracting urban/suburban infrastructure and socio-economic attributes from remote sen.& data are presented. The goal is to relate the user information requirements with the current and proposed remote sensing systems to determine if there are substantive gaps in capability. Several remote sensing systems currently provide some of the desired urban/suburban infrastructure and socio-economic information when the required spatial resolution is poorer than 4 by 4 m and the temporal resolution is between 1 and 55 days (e.g., Landsat MSS and Thematic Mapper, SPOTI-4, Russian TK-350, RADARSAT, Indian IRS-ICD, NOAA AVHRR, GOES, Meteosat). Current high spatial resolution sensor systems such as the Russian SPIN-2 KVR-1000 (2-by 2-m panchromatic; when in orbit) and proposed sensor systems (EOSAT Space Imaging IKONOS 1-by 1-m panchromatic; Earth Watch Quickbird 0.82 by 0.82 m; Orbview-3 1 by I m) may provide additional capability. Large-scale metric aerial photography or digital camera imagery with spatial resolutions ranging from I 0.25 to 1 m will still be required to satisfy several important urban/suburban information requirements. TABLE 1.

This study introduces change detection based on object/neighbourhood correlation image analysis and image segmentation techniques. The correlation image analysis is based on the fact that pairs of brightness values from the same geographic area (e.g. an object) between bi-temporal image datasets tend to be highly correlated when little change occurres, and uncorrelated when change occurs. Five different change detection methods were investigated to determine how new contextual features could improve change classification results, and if an object-based approach could improve change classification when compared with per-pixel analysis. The five methods examined include (1) object-based change classification incorporating object correlation images (OCIs), (2) objectbased change classification incorporating neighbourhood correlation images (NCIs), (3) object-based change classification without contextual features, (4) per-pixel change classification incorporating NCIs, and (5) traditional per-pixel change classification using only bi-temporal image data. Two different classification algorithms (i.e. a machine-learning decision tree and nearestneighbour) were also investigated. Comparison between the OCI and the NCI variables was evaluated. Object-based change classifications incorporating the OCIs or the NCIs produced more accurate change detection classes (Kappa approximated 90%) than other change detection results (Kappa ranged from 80 to 85%).

No abstract

No abstract