Evaluation and Analysis of Environment Satellite(HJ-1) Data about Land Features Classification

To study properties about land features information extraction of HJ-1A CCD1 remote environment monitoring images, this paper selected the east area of Nileke forest farm in western Tianshan mountains of China as the study area, used maximum-likelihood classifier, Mahalanobis distance classifier, minimum distance classifier and K-means classifier to classify, compare and analyze at two different scales with Landsat5 TM images. The experimental results show the following three phenomena: (1) different classification scales have different accuracies. In the first land use classification system, the classification accuracy of HJ-1A CCD1 images are lower than TM images, but higher in the second forest land classification system. (2) Accuracy results of maximum-likelihood classification show that it is the best algorithm to classify land use types. In the first land use classification system, TM total accuracy is up to 85.1% and Kappa coefficient is 0.8. In the second land use classification system, the result is up to 85.4% and kappa coefficient is 0.74. and(3) judgment both from the view of visual interpretation and quantitative accuracy testing, unsupervised classification method with K-means classifier has low qualities in where many land features have characters of scattered distribution and small different spectrum information.