本文基于ＤＳＡ图象中由于选影剂的影响存在灰度差别的特点，提出了一种新的匹配算法─—排除模板迭代算法，成功地实现了ＤＳＡ图象中的运动伪象和灰度伪象的校正。在我们的算法中利用排除模板将存在显著灰度差别的ＤＳＡ序列图象转换为可以利用古典匹配算法的序列图象，而且在用排除模板进行匹配的过程中，逐次选代，直至最佳匹配状态。整个算法实现了全自动，且匹配速度快，搜索范围大，精度可达１个象素点。实验结果表明，此方法对提高ＤＳＡ成象质量很有作用。 In this paper, a new matching algorithm based on the difference of gray level in the DSA image is proposed, which excludes the template iteration algorithm and successfully implements the motion artifact and the gray pseudo-artifact in the DSA image Correction of the icon. In our algorithm, an exclusion template is used to convert DSA sequence images with significant gray-level differences into sequence images that can utilize the classical matching algorithm, and during the elimination of matching templates, successive substitutions are made until the best match status. The entire algorithm is fully automatic, and the matching speed, search range, accuracy of up to 1 pixel. Experimental results show that this method is very useful for improving DSA imaging quality.