Zhang Yangzhen, Zhang Shuwen, Chang Liping, Zhang Bai, Chen Chun
The maize yield estimation is to monitor maize growing in real-time;and estimate its area and yield dynamically through obtaining image feature and spectral data of environment of maize growing by means of remote sensing method, combined with agronomy method. Lishu County in the northeast is a base of commodity grain in China. The area of maize planting accounts for 70 percent of cereal. According to matural condition and lanuse,the whole area is divided into 3 regions of maize yield estimation and 6 types. The maize yield estimation has been carried out under the support of remote sensing technology system,which includes 4 subsystems,i,e. information of maize yield estimation,basic natural envoronment subsystem,remote sensing image analysis,and models and measurement data. GIS was established under the environment of 486micro-computer and ARC/INFO and DBASE Ⅲ. Remote sensing image was analyzed by IS2 S600 system in VAX 11/750 mini-computer. Remote sensing image processing is a key step in technology system for estimation of maize yield,which mainly includes two parts: (1)information extraction and area estimation of maize in the different growing conditions (2)monitor-ing and dividing of maize growing by calculation of greeness indexes. The maize yield was estimated by both remote sensing survey method and remote sensing model analysis. The former has resulted in maize area data of the whole county,combining unit yield gained by traditional agronomy method, and the estimated total yield is 1 150 000 tons, close to the real yield 1 130 000 tons, The latter has established a mathematic model suitable for the regions of high and mid-yield in Lishu County on the basis of the regional division.