Study on auto-extraction of winter wheat planting area from TM image based on pattern recognition technology

This work is a part of yield estimation of main crops by remote sensing technology and it has made important progress. Quick and precise extraction of planting area is a key problem in the research of estimation of crops yield by remote sensing technology. The focus of this paper is on extracting pure wheat land and intercropping wheat land automatically based on the selected proper TM information source on which multidimensional greenness map was built and characteristics and development of winter wheat and pattern recognition technology. The deviation between the wheat planting area obtained from computer interpretation and manual interpretation respectively is small and meets practical requirements.