基于遗传自适应的维纳滤波图像去模糊算法

发布时间：2018-03-16 21:45

本文选题：维纳滤波　切入点：遗传算法　出处：《广西师范大学学报(自然科学版)》2017年04期 　论文类型：期刊论文

【摘要】：为了去除图像在拍摄过程中由于相机的抖动或物体本身运动产生的匀速直线运动模糊,在维纳滤波图像复原的基础上,结合遗传算法,本文提出一种基于遗传自适应的维纳滤波去模糊算法。此算法用遗传算法自适应优化维纳滤波的正则项k值,可达到所选自适应度函数的最佳图像复原效果。实验结果表明,该算法能够通过选择合适的自适应度函数,尽可能地去除实拍图像的运动模糊,还原图像性能优于维纳滤波算法,具有一定的实际意义。
[Abstract]:In order to remove the blur of the uniform linear motion caused by the camera jitter or the motion of the object itself during the shooting process, the Wiener filter image restoration is combined with genetic algorithm. In this paper, a genetic adaptive Wiener filter de-blurring algorithm is proposed. The genetic algorithm adaptively optimizes the regular term k value of Wiener filter, which can achieve the best image restoration effect of the selected adaptive degree function. This algorithm can remove the motion blur of the real image as much as possible by selecting the appropriate adaptive function, and the performance of the restored image is better than that of the Wiener filter, which has certain practical significance.
【作者单位】：广西师范大学电子工程学院;
【基金】：国家自然科学基金(21327007) 广西研究生教育创新计划(YCSZ2015101)
【分类号】：TP18;TP391.41

【相似文献】