Study on Automatic Extraction of Corn Fields Information on Remotely Sensed Imagery Based on Multi-characters Space

Many researches have focused on the automatic extraction of thematic information from Landsat/TM remotely sensed imagery. A method was put forward for automatic thematic information extraction based on the multi-characters space in remote sensing images. According to the analysis of the result of classification, it was concluded that the new method adopted in the paper could improve the efficiency of thematic information extraction from remotely sensed images; and then supervised classification was adopted in the Landsat/TM image classification, corn land in the study area was extracted from the Landsat/TM images with the precision of up to 85.5%. The extraction of corn fields in the study area from the images was performed again on the basis of the expert database, and it was found that the interpretation was notably improved with the precision of 92.9%. Comparing this classification result with the traditional visual interpretation, it was concluded that the new method adopted in the paper could improve efficiency of thematic information extraction from the remotely sensed images. The new method was also theoretically significant with providing new thoughts for intellectualized interpretation of remote-sensing images.