Techniques for developing land-use classification using moderate resolution imaging spectroradiometer imagery

Abstract. Using NOAA AVHRR or MODIS imagery to create land-use classifications has been attempted for many years. Unfortunately, most of these classifications do not differentiate crop types. Crop models require that vegetation characteristics extracted from an image be the correct crop type. This study compares four techniques to create land-use classifications using MODIS data. These classifications were compared to the ground data that had been set aside, to a mask where each MODIS sized pixel were at least 80% of a single land-used based on a Landsat TM classification for the same year, and to the Landsat TM classification. Using a decision tree method and comparing the classification to an 80% mask resulted in an accuracy of 73% which was the highest accuracy obtained in this study. The study showed that accuracies could range from 37% to 73% depending on the classification process and if segment data, an 80% mask, or a Landsat TM classification were used for accuracy assessment.

[1]  Peng Gong,et al.  Land cover assessment with MODIS imagery in southern African Miombo ecosystems , 2005 .

[2]  STATE-LEVEL CROP MAPPING IN THE U.S. CENTRAL GREAT PLAINS AGROECOSYSTEM USING MODIS 250-METER NDVI DATA , 2005 .

[4]  Kyung-Soo Han,et al.  A land cover classification product over France at 1 km resolution using SPOT4/VEGETATION data , 2004 .

[5]  J. Townshend,et al.  Global land cover classi(cid:142) cation at 1 km spatial resolution using a classi(cid:142) cation tree approach , 2004 .

[6]  R. DeFries,et al.  Classification trees: an alternative to traditional land cover classifiers , 1996 .

[7]  Deriving Land Cover Information over Siberia: Comparing Results from MERIS and MODIS , 2005 .

[8]  Christopher O. Justice,et al.  Special issue on the moderate resolution imaging spectroradiometer (MODIS): a new generation of land surface monitoring , 2002 .

[9]  Philip H. Swain,et al.  Pattern recognition: A basis for remote sensing data analysis , 1972 .

[10]  Julius T. Tou,et al.  Pattern Recognition Principles , 1974 .

[11]  DISCRIMINATING CROPPING PATTERNS FOR THE U.S. CENTRAL GREAT PLAINS REGION USING TIME-SERIES MODIS 250-METER NDVI DATA - PRELIMINARY RESULTS , 2002 .

[12]  Christopher O. Justice,et al.  Towards operational monitoring of terrestrial systems by moderate-resolution remote sensing , 2002 .

[13]  Russell G. Congalton,et al.  A review of assessing the accuracy of classifications of remotely sensed data , 1991 .

[14]  J. Carreiras,et al.  Assessing the extent of agriculture/pasture and secondary succession forest in the Brazilian Legal Amazon using SPOT VEGETATION data , 2006 .

[15]  J. Townshend,et al.  NDVI-derived land cover classifications at a global scale , 1994 .

[16]  Alfredo Huete,et al.  Assessing the seasonal dynamics of the Brazilian Cerrado vegetation through the use of spectral vegetation indices , 2004 .

[17]  Rick Mueller,et al.  Creating a Cropland Data Layer for an Entire State , 2002 .

[18]  Prasad S. Thenkabail,et al.  Ganges and Indus river basin land use/land cover (LULC) and irrigated area mapping using continuous streams of MODIS data , 2005 .

[19]  J. Townshend,et al.  Global discrimination of land cover types from metrics derived from AVHRR pathfinder data , 1995 .

[20]  B. Wardlow,et al.  Analysis of time-series MODIS 250 m vegetation index data for crop classification in the U.S. Central Great Plains , 2007 .