Object-based image analysis (OBIA) provides a better solution for information extraction from high spatial resolution remote sensing image. Currently, selection of scale parameters is often dependent on subjective trial-and-error methods or post-evaluation of multi-segmentation, which directly reduces efficiency of land cover classification. This paper proposes a OBIA classification method combining spatial statistics based adaptive scale parameter pre-estimation and Mean-shift segmentation. Series of object based classification were employed to verify the validity of this method. Experimental results show that the pre-estimated scale parameter can guarantee a classification result with both high classification accuracy and good completeness for land cover classification. This presented method avoids the time-consuming trial-and-error practice so that it speeds up the object-oriented classification procedure.