关键词: 运动模糊, 模糊方向, 局部标准差, 方向微分, 双线性插值

Abstract: Concerning the shortcomings of big recognition error and bad real-time calculation by the minimum differential directional algorithm, a method in combination with local standard deviation and directional derivation was put forward to identify the blurred direction of motion blur image. Firstly, the motion blur image was filtered by local standard deviation to enhance texture details in blurred direction. Secondly, the minimum directional derivation summation was obtained by bilinear interpolation, and its corresponding direction was the blurred direction. Meanwhile, the inherent law was found after concluding and analyzing the minimum directional derivation summation curve. Based on this law, the method of search range decreasing in half was presented to search minimum, which could reduce searching times. The simulation results show that this algorithm not only has high precision and strong immunity, but also meets the requirement of real-time calculation.

Key words: motion blur, blurred direction, local standard deviation, directional derivation, bilinear interpolation