【摘要】 针对经典灰度相关匹配算法耗时多的缺点,提出一种减少计算次数、提高匹配效率的新方法。该方法是先构造一幅相关图像,然后计算它与基准子图像的互相关数据并利用其符号分离候选匹配点,最后在匹配过程中,根据实时图像与相关图像的相关系数符号来选定匹配点区域,从而减少匹配计算次数,提高匹配速度。理论分析和实验表明,该方法不但能够保持经典灰度相关匹配算法具有较高的匹配精度和较强的抗噪能力的特点,而且大大缩短了匹配时间,使得经典灰度相关算法能够在实际中得以应用。更多还原

【Abstract】 To overcomes the weakness of classic gray correlation matching algorithm taking too much times,a new method to decrease the time of computation and raise the matching efficiency was proposed.First,this method constructed a related image.Second,the datum of correlation coefficient between sub-reference image and related image was computed.And then, potential matching points was separated by utilizing the sign of the above step.At last,possible matching point areas were selected by means of the sign of correlation coefficient between related image and real-time image.Theories analysis and expe- riment result suggest that this method can not only keep higher matching accuracy and stronger anti-noise ability of the classic gray correlation matching algorithm,but also reduce the matching time in large quantity,which make it suitable for real-time application.更多还原