Segmentation of multi-temporal ERS-1 SAR imagery
SAR segmentation is an important step for the operational use of satellite SAR imagery in routine mapping exercises. The use of multi-temporal SAR imagery in this respect is of specific interest in areas where optical data are difficult to obtain, due to prevailing weather conditions. This paper presents the results of the application of a generic segmentation method of multi-temporal ERS-1 SAR imagery of the Dutch Flevoland agricultural area. The data were recorded during Autumn 1991, and constitute a series of seven co-registered ERS-1 PRI images. Before segmentation, the data are speckle filtered and then byte- scaled to allow the segmentation of any combination of temporal channels. The various channel combinations are then evaluated with respect to segmentation efficiencies. The results are compared to an existing database of fixed field boundaries and a vector map of 1991 field boundaries derived from optical data (SPOT). One of the final objectives is to generate automatically multi-temporal backscattering signatures (field averaged PRI data extracted with polygons generated in the segmentation procedure) for the training of both supervised classification by means of neural networks, and supervised tillage monitoring. In particular, the potential to significantly advance the time of earliest estimates of crop acreage, by combining results from segmentation and knowledge based classification, is of interest in this framework. For areas where timely optical data are available, a hybrid approach can be adopted, still using the same segmentation algorithm.
Bibliographic Reference: Paper presented: European Symposium on Satellite Remote Sensing II, Paris (FR), September 25-28, 1995
Availability: Available from (1) as Paper EN 39293 ORA
Record Number: 199511455 / Last updated on: 1995-11-03
Original language: en
Available languages: en