An evaluation of fuzzy classifications from IRS 1C LISS III imagery: A case study

Remote sensing data are frequently used to produce crisp and fuzzy classifications for land cover applications. Fuzzy approaches are attractive for classification of images dominated by mixed pixels. Recently, fully-fuzzy classification that can incorporate mixed pixels in all the three stages of a supervised classification has been recommended. In this Letter, the results of a case study on fully-fuzzy classification of Indian Remote Sensing (IRS) 1C Linear Imaging Self Scanning Sensor (LISS) III imagery are reported. The results illustrate that fully-fuzzy classification produces more accurate land-cover mapping than the conventional crisp classification. For instance, the areal extents of three dominant classes (i.e. agriculture, forest and grass) obtained from fully-fuzzy classification differ by only 13% from the actual areal extents, compared to 34% difference in area observed from crisp classification.

[1]  N. Campbell,et al.  Derivation and applications of probabilistic measures of class membership from the maximum-likelihood classification , 1992 .

[2]  Elisabetta Binaghi,et al.  A fuzzy set-based accuracy assessment of soft classification , 1999, Pattern Recognit. Lett..

[3]  G. M. Foody,et al.  Relating the land-cover composition of mixed pixels to artificial neural network classification outpout , 1996 .

[4]  J. Settle,et al.  Linear mixing and the estimation of ground cover proportions , 1993 .

[5]  Giles M. Foody,et al.  Approaches for the production and evaluation of fuzzy land cover classifications from remotely-sensed data , 1996 .

[6]  Fangju Wang,et al.  Fuzzy supervised classification of remote sensing images , 1990 .

[7]  Giles M. Foody,et al.  Cross-entropy for the evaluation of the accuracy of a fuzzy land cover classification with fuzzy ground data , 1995 .

[8]  Martin Brown,et al.  Linear spectral mixture models and support vector machines for remote sensing , 2000, IEEE Trans. Geosci. Remote. Sens..

[9]  Giles M. Foody,et al.  Estimation of sub-pixel land cover composition in the presence of untrained classes , 2000 .

[10]  Giles M. Foody,et al.  Mapping Land Cover from Remotely Sensed Data with a Softened Feedforward Neural Network Classification , 2000, J. Intell. Robotic Syst..

[11]  Giles M. Foody,et al.  Fully-fuzzy supervised classification of sub-urban land cover from remotely sensed imagery: Statistical and artificial neural network approaches , 2001 .

[12]  J. Bezdek,et al.  FCM: The fuzzy c-means clustering algorithm , 1984 .

[13]  B. Madhavan,et al.  Integration of IRS-1A L2 data by fuzzy logic approaches for landuse classification , 2000 .

[14]  G. Foody,et al.  Sub-pixel land cover composition estimation using a linear mixture model and fuzzy membership functions , 1994 .

[15]  F. Maselli,et al.  Use of probability entropy for the estimation and graphical representation of the accuracy of maximum likelihood classifications , 1994 .

[16]  Roger G. Barry,et al.  Cloud classification from satellite data using a fuzzy sets algorithm - A polar example , 1989 .

[17]  C. Woodcock,et al.  Theory and methods for accuracy assessment of thematic maps using fuzzy sets , 1994 .