Detection Of Seasonal And Long-term Changes In Land Cover From Multi Temporal Landsat MSS Data
説明
Procedures were developed and tested for detecting changes in land cover by using multitemporal Landsat MSS data. Our main concern Is to detect the occurrences of seasonal changes and long-term changes in land cover. Four observation dates by Landsat were selected in spring and winter at an interval of 7 or 8 years. Four bands of MSS were used for each date. CCT values were transformed to the values comparable to spectral reflectances by using references. The change detection.procedures proposed here use no training data, and are pre-classification methods. In one procedure spectral differences are calculated at each pixcl between every two dates out of four by using an Euclidean distance norm of the four band data. From these norms the spectral changes among the four dates are checked in every pixel. and seasonal and long-term changes are detected. Principal component procedure is also tested.
収録刊行物
-
- International Geoscience and Remote Sensing Symposium, 'Remote Sensing: Moving Toward the 21st Century'.
-
International Geoscience and Remote Sensing Symposium, 'Remote Sensing: Moving Toward the 21st Century'. 1 215-216, 2005-08-24
IEEE
