伪随机编码图像的特征点自动检测方法

Automatic feature corner detection in pseudo-random encoded image

针对伪随机阵列编码的彩色结构光系统,提出了一种角点自动检测、识别方法。它首先利用聚类分析算法,确定摄像机图像角点和编码模板图像的特征点匹配关系,实现编码图像的识码、解码;然后获取编码图像中有效的、非冗余的角点,并自动输出角点的行列数目及其位置坐标,为亚像素角点检测提供了充分的输入数据。最后,将该算法应用于自主开发的彩色结构光系统上,并用实验检验,结果表明该方法准确、有效,有助于结构光系统的自动化标定。

In the pseudo-random coded structured light system,the automatic corner detection method is proposed.First,the data cluster method is used to determine the correspondence between the camera image and the projector image,it solves the pattern codification;and then by the corner detection method,the coordinates of all the proper and non-redundant corners in the encoded image are given automatically,as well as the row and the column numbers of these corners.These provide the important data for the sub-pixel location of corners.The method is applied to the system that authors develop,experimental results demonstrate that this method is effective for solving the corner detection and correspondence problem in the course of projector calibration.