Full polarimetric synthetic aperture radar (SAR) possessed rich polarization information, it has a significant advantage for coverings recognition. A fusion method which took into account polarization characteristics of full polarimetric SAR is proposed based on SAR and medium resolution optical image. Full polarimetric SAR is carried out polarimetric target decomposition, polarization characteristics and optical image is fused with the improved hue, saturation, value (HSV) transform method. The fusion image is classified based on the object-oriented method. The results show that the proposed fusion method is superior to the traditional HSV fusion method for effectively using the polarimetric information and texture information of full polarimetric SAR. Object-oriented classification method can reduce the speckle noise of fusion image from SAR. The overall classification accuracy is better than that of high resolution optical image, and the classification accuracy of coverings which is sensitive to the polarization information is obviously better than that of high resolution optical image.