In speckle interferometry, there are a large number of speckle noises in the speckle phase map, which need to be filtered. However, the existing filtering evaluation indexes are not suitable to determine the continuous filtering process of single image, and the automation of the filtering process is difficult to be achieved. Therefore, a filtering evaluation method based on smoothing spline fitting is presented in this paper. Firstly the smoothing spline fitting method is used to process the distribution of the phase map . Secondly, the root-mean-square error is used to evaluate the result of continuous filtering process by the smoothing spline fitting. Finally, the experimental images of speckle interferometry are used to verify the proposed method. The results of the experiment show that the method can realize the quantitative analysis of the filtering effect and the judgment of the completion of the image filtering in continuous filtering process, and the effectiveness and universality of the method are also verified.