Beijing-1 small satellite was launched Oct.27 2005 and has taken part in the plan of China high-performance earth observation after finishing on-orbit test period. Two kinds of sensors were carried on the satellite. One is 3-band multi-spectral senor whose spatial resolution was 32m, the other panchromatic sensor whose spatial resolution was 4m. In order to ensure truly utility for small satellite data, preliminary deep processing system had been developed for receiving, preprocessing, and data-distribution. Meanwhile, several key questions must be deal with including radiometric calibration, geometric precise rectification, orthographic rectification, image fusion and application demonstration. The paper will focus on the works of the second part including RPC orthographic rectification model and how to optimize algorithms of orthographic rectification which consider the feature of 4m high spatial resolution. RFM is a generalized sensor model, which uses RPC parameters to perform orthographic rectification in no need of orbit parameters and sensor imaging parameters. It is independent on sensors or platforms and supports any object space coordinate system with a variable coordinate system. Compared to linear transformation and polynomial transform, RFM has the highest positioning accuracy. Because RPC is determined by applying the least squares principle to GCP data, approximate error can be evenly distributed through RFM rectification. Based on the experiment on the Beijing-1 high resolution small satellite data using RFM and improved RFM, a generalized model of orthographic rectification of high resolution small satellite data can be developed. The experiment proves: Using second-order improved RFM to rectify the Beijing-1 small satellite image has a sub-pixel positioning accuracy that is close to the accuracy of the rigorous sensor model based on the collinearity equation when the GCPs are evenly distributed.