摘要
在手机玻璃盖板的缺陷检测中,为了获得缺陷特征明显的高精度图像,不仅需要搭建高质量图像采集系统,而且需要对线阵相机进行精密标定.结合工业现场需求,本文提出了一种基于改进遗传优化算法的线阵相机标定方法.在已有线性扫描模型的基础上,引入线阵扫描图片图像畸变矫正,推导了新的标定参数.在已有无畸变矫正的线性解析解基础上,引入遗传算法进行新参数的求解优化,实现了新标定参数的非线性求解.实验结果表明,与不加入图像畸变矫正的求解方法相比,所提方法具有更高的测量精度.
In order to obtain high-precision image with obvious defect characteristics in defect detection of mobile phone glass cover,the linear array camera with high-quality image need to be calibrated precisely.A calibration method with improved genetic optimization algorithm of linear array camera was proposed.Based on the existing linear scan model,the image distortion of linear scan image was corrected and the new calibration parameters were derived.The solution of new calibration parameters was a non-linear problem.An improved genetic optimization algorithm was proposed to optimize the solution process.The experimental results show that the proposed method has higher calibration accuracy in the actual scanning detection environment than the general genetic algorithm and image distortion correction.
作者
王向周
陈冬清
郑戍华
乔亚军
WANG Xiang-zhou;CHEN Dong-qing;ZHENG Shu-hua;QIAO Ya-jun(School of Automation, Beijing Institute of Technology,Beijing 100081, China)
出处
《北京理工大学学报》
EI
CAS
CSCD
北大核心
2020年第8期861-866,共6页
Transactions of Beijing Institute of Technology
关键词
线阵相机
视觉标定
畸变矫正
遗传优化
linear array camera
visual calibration
distortion correction
genetic optimization
