Abstract: The paper aims to study the application of high resolution remote sensing satellite image to the investigation, monitoring and statistics of arid sand vegetation. The author took Qingtu Lake located in the downstream of Shiyang River in Minqin County of Gansu Province as the study object, used the French Pleiades-1 satellite image data, and adopted the object oriented classification method, to extract and classify information of vegetation in desert area and evaluate the accuracy of the classification results. The results showed that the overall classification accuracy of object-oriented information extraction reached 95%, Kappa coefficient reached 0.8035. The classification results had high accuracy, effectively avoided the noise and spectral image based on pixel classification method, and had very strong practicability, providing a new idea and method for the high resolution satellite image classification.