Crop classification accuracy as influenced by training strategy, data transformation and spatial resolution of data
- 35 Downloads
The possibility of improving classification accuracies using different training strategies and data transformations within the framework of a supervised maximum likelihood classification scheme was explored in this study. The effect of spatial resolution of data on the accuracy of classification was also studied Single-pixel training strategy resulted in improved classification accuracy over the block-training method. Data transformations gave no significant improvements in accuracy over untransformed data. There was a reduction in classification accuracy as resolution of data improved from 72 m (LISS I) to 36 m (LISS II) while other sensor characteristics remained same.
KeywordsRemote Sensing Classification Accuracy Kappa Coefficient Data Transformation Training Strategy
Unable to display preview. Download preview PDF.
- Campbell J B (1981). Spatial correlation effects upon accuracy of supervised classification of land cover, Photogramm. Eng. Remote Sensing, 47(3): 355–363.Google Scholar
- Chuveico E and Congalton R G (1988). Using clustering analysis to improve the selection of training statistics in classifying remotely sensed data, Photogramm. Eng. Remote Sensing, 54(9): 1275–1281.Google Scholar
- Craig R G (1979). Autocorrelation in Landsat data, In Proc. 13th Int. Symp. Remote Sensing Environ., ERIM (Ann Arbor. Michigan), pp. 1517–1524.Google Scholar
- Dadhwal V K and Parihar J S (1985). Estimation of 1983–84 wheat acreage of Karnal district (Haryana) using Landsat MSS digital data, Scientific Note, IRSUP/SAC/CPF/SN/09/85.Google Scholar
- Gong P and Howarth P J (1990). An assessment of some factors influencing multispectral land cover classification, Photogramm. Engg. Remote Sensing, 56(5): 597–603.Google Scholar
- Hixson M, Scholz D, Fuhs N and Akiyama T (1980). Evaluation of several schemes for classification of remotely sensed data, Photogramm. Eng. Remote Sensing, 46(12): 1547–1553.Google Scholar
- Medhavy T T, Dadhwal V K, Parihar J S, Ruhal D S, Khera A P, Jarwal S D and Singh J (1989). Use of single acquisition IRS LISS I digital data for preharvest wheat acreage (1988–89) estimation in Haryana, Scientific Note, RSAM/SAC/CAPE/SN/18/89.Google Scholar
- Ruhal D S, Dadhwal V K, Parihar J S, Jarwal S D, Medhavy T T, Singh J and Khera A P (1988). Studies on within crop variability in wheat using Landsat TM data, In Proc. National Symp. Remote Sensing in Rural Development, Nov. 17–19, 1988, Hisar, pp. 199–208.Google Scholar
- Ruhal D S, Medhavy T T, Jarwal S D, Dadhwal V K, Khera A P and Parihar J S (1990). District-level wheat acreage estimation in Haryana using IRS LISS I digital data, In Proc. National Symp. Remote Sensing for Agricultural Applications, Dec. 6–8, New Delhi, pp. 250–255.Google Scholar
- Shimoda H and Sakata T (1988). Accuracy of land use classification for SPOT image data, SPOT-1 Image Utilisation, Assessment Results, Cepadues Edition, Toulouse, France, pp. 631–636.Google Scholar
- Williams D L, Irons J R, Markham B L, Nelson R F and Toll D L (1983). Impact of TM sensor characteristics on classification accuracy, In Proc. IGARSS '83, San Francisco, California, pp. PS 1 5.1–5.9.Google Scholar