CHANGE DETECTION IN BUILD-UP AREAS ON THE BASIS OF STRUCTURAL FEATURES OF SATELLITE IMAGES
S. M. Borzov1, A. O. Potaturkin1, O. I. Potaturkin1,2
1Institute of Automation and Electrometry, Siberian Branch, Russian Academy of Sciences, pr. Akademika Koptyuga 1, Novosibirsk, 630090 Russia
2Novosibirsk State University, ul. Pirogova 2, Novosibirsk, 630090 Russia
Keywords: дистанционное зондирование Земли, выделение антропогенных территорий, обнаружение изменений, спектральные и пространственные (текстурные, структурные) признаки, классификация изображений, remote sensing, selection of man-made areas, change detection, spectral and spatial (textural and structural) features, classification of images
Abstract
It is shown that spatial structural features based on determining the maximum angle within which each pixel has a neighborhood that is not shadowed by other pixels of the contour provide the best efficiency in detecting man-made areas in satellite images of moderate resolution. A combined method of detection of build-up area expansion on the basis of multispectral data provided by the SPOT-4 satellite is proposed and demonstrated. This method implies the formation (with allowance for chosen features) of spatial masks of man-made objects in the reference (earlier taken) image and natural zones of the tested (later taken) image, searching for spectral-temporal anomalies, and sequential masking of selected pixels with the use of masks being formed.
|
