模糊目标轮廓图像分割研究

Image segmentation research on blurred object contour

针对实际图像成像过程中,由于图像传感器件的特性和光学衍射效应造成的目标轮廓模糊,形成了斜坡边缘。分析了模糊轮廓的边缘特性,研究了不同算子对阶跃边缘和斜坡边缘的影响,提出了模糊引起图像灰度的内插平滑,降低了轮廓与周围点灰度的差异性,产生了过渡带。研究了模糊对灰度值的影响,提出了灰度乘方的梯度边缘检测算法,增强了过渡区域灰度变化的差异性,相应减小了相机采集过程带来的图像轮廓模糊。通过实验表明,该算法有效地解决了模糊目标轮廓的分割问题。

During the imaging process of the actual imaging,due to the characteristics of image sensors and the fuzzy target coutour of optical diffraction effect,a ramp edge is formed.The edge characteristics of the blurred contour is analyzed,and the effects of different operators on the step edge and slope is studied.The smooth interpolation of gray level image caused by blur is put forward,which reduces the gray differences between the contour point and the neighborhood point meanwhile,transitional zone is produced.The effect of blur on gray value is researched and the gray power based on the gradient edge detection algorithm is proposed,which enhances gray-scale changes differences in the transition zone,and decreases the image contour blur brought by the camera acquisition correspondingly.Experiments show that the algorithm can effectively solve the image contour segmentation of blurred objectives.