在获取图像的过程中，由于被摄物与成像系统产生相对运动，使得图像的像素并不代表该点的实际灰度，而是一定范围内相邻像素灰度值的叠加。为了有效地恢复运动模糊图像，从像素运动的角度分析了因运动造成的图像模糊的实际过程，并首先揭示了运动造成图像降质的内在原因，然后根据模糊图像像素间的关系建立了一维和二维由匀速直线运动所造成的模糊图像的模糊模型，并依据图像降质过程的逆过程推导出恢复模型。为验证该模糊图像恢复技术的效果，对匀速直线运动造成的图像模糊进行了模拟和恢复实验，取得了令人满意的效果，尤其对字母的恢复，效果更好。另外，对边界像素问题还进行了专门的讨论，同时对恢复过程中产生误差的原因也进行了分析，最后总结了该算法的优越性和需要进一步改进的问题。

In acquiring image, due to the relative motion between an imaging system and its objects, the every value of grey scale of pixels expresses several ones within area instead of itself one pixel. It, thereby, results in imaging degrading, and engenders blurring imaging. The paper analyses a real process about image blurred as a result of object motion, reveals an intrinsical cause about image-degraded owing to object motion, and establishes blurring-model by which one dimension and two dimensions uniform velocity rectilinear motion result in image blurred abide. Based on result of blurring imaging pixel motion, the writers introduce a restoring model of blurring imaging from reverse process of degrading-process and conclude a restoring technique of motion blurring imaging from it. As an instance, the paper models a blurring process for uniform velocity rectilinear motion and restores it again. The example conforms the technique is available, especially for restoring characters and texts. A special discuss on boundary pixel problem is given. An analyse about error cause in process of restoring is introduced. In the end, the method's advantage and disadvantage and how to improve on it were discussed.