改进的运动模糊尺度检测方法  

Improved Detection Method of Motion Blur Scale

作　　者：王鸿雁[1] 崔红霞[1] 刘佳琪[1] 刘畅[1] 

机构地区：[1]渤海大学信息科学与技术学院,辽宁锦州121000

出　　处：《计算机技术与发展》2016年第12期69-72,76,共5页Computer Technology and Development

基　　金：国家自然科学基金资助项目(41371425)

摘　　要：由于运动模糊的普遍存在,导致影像质量不断下降。针对这个问题,提出一种改进的二次微分自相关算法,以更精确地鉴别模糊尺度。采用一阶微分算子将模糊图片进行微分,再用Sobel算子对其进行二次微分,得到梯度图像。之后对梯度图像求自相关,将其结果取平均值并绘制点扩散函数鉴别曲线,通过计算零频尖峰与负尖峰的距离,得出模糊尺度。通过MATLAB对不同图像进行多次仿真实验,证明了算法的可行性。仿真结果表明,改进的方法对模糊尺度的识别准确度高、误差较小,对模糊图片具有更好的适应性和抗噪性。Because of the widespread existence of motion blur,the image quality is constantly falling. To solve this problem,an improved second differential auto correlation algorithm is proposed to identify fuzzy scale more accurately. The fn＇st order differential operators is adopted to carry out the differential of the fuzzy image. And then,the gradient image is obtained by using the Sobel operator to make the two differential. After that, the auto-correlation of gradient image is obtained and on the basis of the average value of the results, the point spread function is drawn. By calculating the distance between the zero frequency and the negative spike, the size of the vague scale is obtained. The feasibility of the algorithm is proved by several simulation experiments on different images through the MATLAB software. The results show that the improved method has a higher accuracy and smaller errors. At the same time ,the adaptability and anti noise capability of the algorithm is higher.

关 键 词：运动模糊 差分自相关 微分算子 频谱估计法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]