Title :
Quantifying the performance of SAR segmentation algorithms
Author :
Caves, Ronald G. ; Quegan, Shaun
Author_Institution :
Centre for Earth Obs. Sci., Sheffield Univ., UK
Abstract :
Objective methods for analysing the performance of SAR segmentation algorithms, in terms of segment homogeneity, are discussed. Tests are based on the segmentations generated by two algorithms, RWSEG and ANNEAL, of an ERS-1 image of an agricultural scene. Homogeneity is assessed using the coefficient of variation of segments and the ratio of the original and segmented images. The relation between these approaches and the effect of segment size are discussed. The measures of homogeneity which are derived correspond to the results of a visual comparison
Keywords :
geophysical signal processing; geophysical techniques; image segmentation; radar imaging; remote sensing by radar; spaceborne radar; synthetic aperture radar; ANNEAL; ERS-1; RWSEG; SAR imaging; agriculture; geophysical measurement technique; image segmentation algorithm; land surface; performance; radar remote sensing; segment homogeneity; synthetic aperture radar; terrain mapping; vegetation mapping; Algorithm design and analysis; Annealing; Change detection algorithms; Image edge detection; Image segmentation; Iterative algorithms; Layout; Performance analysis; Pixel; Speckle; Testing;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 1995. IGARSS '95. 'Quantitative Remote Sensing for Science and Applications', International
Conference_Location :
Firenze
Print_ISBN :
0-7803-2567-2
DOI :
10.1109/IGARSS.1995.521772
