Abstract: The integration of remote sensing and geographical information system technology is an optimal method for maize yield estimation.In this study,a county agricultural information system has been established,which supports various data capture,information complex and spatial analysis.Based on the system,maize yield estimation mechanism is explored with remote sensing,whereby maize yield factors can be derived from remote sensing indices.Some relative models between vegetation index (PVI,RVI,NDCI,WDVI,TSAVI,etc.) and Leaf Area Index (LAI),chlorophyll concentration and biomass are developed.Crop Water Stress Index (CWSI) is calculated to indicate the stress of maize growing conditions.Therefore,with the aid of the system,PVI,NDCI and CWSI,which respectively stand for maize biological parameters and imitate its ecological environment,are two categories index in a comprehensive yield estimation model.Then,considering the difference of maize growing environment in Lishu county,not only three yield classes of maize are determined but also three yield estimation models for each class are developed.Moreover,yield estimation models for each maize growing stage are discussed in this paper.