提出结合滤波算法的不敏卡尔曼滤波(UKF)相位解缠方法。该方法把UKF、传统路径跟踪策略、全方位的局部相位梯度估计技术以及干涉图小窗口滤波算法有效结合起来,能在相位解缠的同时进行干涉相位噪声抑制,既不受相位残差点影响,又避免了传统方法在相位解缠之前须尽可能滤除干涉图中相位噪声的不足。模拟和实测数据试验结果验证了本文方法的有效性。 An unscented Kalman filter (UKF) phase unwrapping method based on the combined filtering algorithm is proposed. This method effectively combines UKF, traditional path tracking strategy, omni-directional local phase gradient estimation and small interference filter windowing algorithm to suppress interference phase noise while phase unwrapping, and is not affected by phase residuals. But also avoids the traditional method to filter out the phase noise in the interferogram as much as possible before the phase unwrapping. The simulation results and experimental data show that the proposed method is effective.