基于三线结构光的相机平面标定方法研究  

Research on Camera Plane Calibration Based on Three-Wire Structured Light

作　　者：吴芳[1] 茅健[1] 周玉凤[1] 李情[1] 

机构地区：[1]上海工程技术大学机械工程学院,上海201620

出　　处：《计算机测量与控制》2017年第7期206-208,229,共4页Computer Measurement &Control

摘　　要：相机标定技术是结构光三维视觉测量的关键技术之一,结构光测量系统的相机标定的精度对三维测量的精度有很大影响;首先对三线结构光系统图的相机标定方法进行了分析,简单介绍了工业相机成像的几何模型及标定的原理;其次利用Harris角点检测方法提取特征点坐标,并选用了BP神经网络来校正工业相机的畸变模型,以提高标定算法的优化速度和标定精度;最后采用张正友的平面标定法对校正后的摄像机模型进行标定实验,由实验结果知,该方法具有一定的准确性和有效性,在一定误差范围内,基于神经网络畸变校正的张正友相机标定能够有效提高视觉检测的精度。Camera calibration technology is one of the key techniques of structured light 3D vision measurement. The precision of camera calibration of structured light measurement system has great influence on the accuracy of 3D measurement. The camera calibration method of the three-line structured light system diagram is analyzed, and the geometric model and calibration principle of the industrial camera are introduced. Secondly, the Harris corner detection method is used to extract the feature point coordinates, and BP neural network is used to correct The calibration model of the camera is calibrated by Zhang Zhengyou＇s plane calibration method, and the experimental results show that this method has certain accuracy and validity, and it can improve the precision and effect of the calibration algorithm. In a certain error range, the calibration of the camera based on neural network distortion correction can effectively improve the accuracy of visual inspection.

关 键 词：三线结构光 相机标定 针孔模型 畸变模型 平面标定方法 BP神经网络 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]