This paper seek to evaluate the application potential and accuracy evaluation of CBERS-02B high resolution (Pan) data. Taking Xixiang County of Shaanxi Province as a test area，radiometric and color enhancement is performed to HR image; then the HR image was fused with the CCD images based on five algorithms，and image qualities are evaluated between original images and fused images in terms of qualitative and quantitative aspects. Finally, land-use categories were extracted and classification accuracy was evaluated based on the two datasets. The results show that land-use classification accuracy is improved greatly by fusion of 02B high resolution data and CCD data, and Brovey algorithm is best for keeping the spatial textures and spectral characteristics of images among five algorithms. HR image should be applied widely in resources investigations and ecological environment monitoring.