Abstract: The purpose of this paper is to monitor the growth of new planting trees in the cropland being converted to forests in the project of converting cropland to forest using QuickBird high resolution (0.6 m) remote sensing image in Zhangjiakou,Hebei Province. Ortho-rectification has been done for QuickBird image. Gaussian filter was applied to the image to improve the tree crown information and control image noise. Tree crown classifictation of QuickBird data was performed by using object-oriented image analysis,therefore,crown area was obtained and new planting trees survival rate was calculated based on the tree crown classification results. Finally,accuracy inspection had been done according to field measurement data,the results showed that the average error of crown area was 0.763 m²,the accuracy of survivae rate was 89.837%,The study will provide a scientific basis for programming and management of the project of converting cropland to forest.

Key words: QuickBird imagery, convertion of cropland to forest, object-oriented image analysis, tree crown, survival rate