基于改进Chan-Vese模型的液晶显示屏Mura缺陷分割  

LCD Mura defectimage segmentation based on improved Chan-Vese model

作　　者：陈凌海[1] 姚剑敏[1] 郭太良[1] 

机构地区：[1]福州大学物理与信息工程学院,福建福州350002

出　　处：《液晶与显示》2016年第6期613-620,共8页Chinese Journal of Liquid Crystals and Displays

基　　金：国家"863"重大专项(No.2013AA030601)~~

摘　　要：液晶显示屏Mura缺陷是一类较难检测的显示缺陷,它具有对比度低、背景亮度不均匀、边缘模糊等特点。针对传统Chan-Vese模型(C-V模型)对其分割时存在误分割及速度慢的问题,本文提出一种改进的C-V模型。首先,依据曲线演化理论,简化了传统C-V模型的图像数据力驱动项,这样减少了迭代过程中的计算量,提高了分割的速度。其次,为了平衡图像的亮度不均匀,在模型中引入一个新的能量项,该能量项与轮廓曲线内、外部之间的亮度差有关,提高了分割的准确性。最后,在算法的实现过程中引入迭代停止的判别式,通过设定分割的精度可以实现迭代的自动停止,并有利于正确地分割出目标。实验结果表明,本文提出的改进C-V模型能够准确分割背景不均匀的Mura缺陷,并且具有较快的速度。LCD Mura defect is difficult to detect.It has the features of low contrast,uneven brightness,blurry edge,and so on.The traditional Chan-Vese model（C-V model）cannot accurately segment the Mura defect and the segmentation speed is slow.In order to solve the problems,this paper presents an improved C-V model.First of all,according to the theory of curve evolution,the image data force in the traditional C-V model is simplified to reduce the iterative process of calculation and improve the segmentation speed.Secondly,in order to reduce the effect from non-uniformity brightness of image,the model is proved by adding a new energy term that related to the brightness difference between inside and outside area of the contour curve to improve the accuracy of the segmentation.Finally,the iterative stopping criterion is introduced in the implementation process of the algorithm,the setting of the segmentation accuracy can aid both in achieving the automatic iteration stop and segmenting the target correctly.Experimental results indicate that improved C-V model in this paper canaccurately segment the background uneven Mura defect and it has a faster speed.

关 键 词：CHAN-VESE模型 液晶显示屏Mura缺陷 亮度不均匀 

分 类 号：TN141.9[电子电信—物理电子学]