双阈值屏幕显示缺陷检测方法

Dual-threshold screen display defect detection method

机器视觉检测液晶屏幕显示缺陷过程中,因相机系统参数设置、屏幕显示亮度等影响,导致采集到图片存在亮度不均致使缺陷检测精度降低。针对这一问题,本文提出了一种基于模块的双阈值缺陷检测算法。首先,将无缺陷屏幕作为标定样本,并对其按空间进行模块划分;其次,对划分后模块区域参数进行统计,以此为依据设置检测阈值;最后,将待检测屏幕图像划分为模块,并根据对应模块区域阈值判断是否存在缺陷。此外,本文设计了一种缺陷检测系统,对上述方法进行了实验验证,结果表明:该方法能有效提高对亮度显示不均匀图片的缺陷检测精度。

In the process of machine vision detection of LCD screen display defects,due to the influence of camera system parameter setting and screen display brightness,the collected pictures have uneven brightness,resulting in precision of defect detection reduced.Aiming at this problem,a module-based dual-threshold defect detection algorithm is proposed.Firstly,the defect-free screen is used as the calibration sample,and it is divided into modules by space.Secondly,the parameters of the divided module area are counted,and the detection threshold is set based on this.Finally,the screen image to be detected is divided into modules,and judge whether the defects exist or not according to the threshold of the corresponding module area.In addition,a defect detection system is designed,and the above method is verified by experiments.The results show that the method can effectively improve the defect detection precision of uneven brightness images.