确定对象单目视觉位姿检测算法的研究

Study on Pose Detection Algorithm for Specific Object Based on Monocular Vision

针对确定作业对象的位姿检测,提出一种基于Hausdorff距离的单目视觉检测算法,即首先依据建立的数学模型构建作业对象在虚拟摄像机下成像的二维模板序列;接着在图像匹配时为降低计算量及加快图像匹配速度,以部分Hausdorff距离作为匹配测度,采用阈值法先行的分层搜索策略进行检测。通过对不同位姿下的夹紧块影像进行检测实验,结果证明了该算法的有效性和快速性。

This paper brought forward a monocular vision detection algorithm in allusion to a specific object＇s pose based on Hausdorff Distance.At first,according to the established mathematic model,a 2D template sequence was generated by projecting the specific object in different poses into the image plane of a virtual camera.Then,in order to predigest calculations and accelerate the matching speed during the image matching,the algorithm adopted the local-mean Hausdorff Distance as matching estimation and adopted the search strategy based on hiberarchy which reduced the searching rang by the threshold method before accurate match.The experiments that measured the poses of the clamp via the different clamp images respectively were given to testify the validity and speediness of this algorithm.