Abstract: The imaging of dynamic object with rotating parts cannot be used for feature extraction directly. To get contour feature of the target, an effective feature extraction method is proposed. Firstly, the adaptive wavelet transform is performed to separate the signal parameters of moving parts from the target's. Then, based on the discontinuity point of image gray scale, the target is separated from the background. Finally, an adaptive boundary convergence method is used to extract the contour features of the target region, and a good contour feature is obtained. The effectiveness and stability of the proposed algorithm is verified by measured data.