CAJ | 학술논문

Harris角点检测算法是一种经典算法，但对于大尺度图像，误检现象比较严重，并且耗时过长。本文提出一种新的检测算法，通过对圆形区域进行非极大值抑制，能够明显地降低角点检测时间，并且能够有效地减少误检。同时，在不改变任何参数的情况下，对于图像旋转，能够明显地减小提取的差异。通过实验对比，本算法简单方便，提取角点也非常有效，相对于传统Harris算法具有更好的角点检测性能。
Harris각점검측산법시일충경전산법，단대우대척도도상，오검현상비교엄중，병차모시과장。본문제출일충신적검측산법，통과대원형구역진행비겁대치억제，능구명현지강저각점검측시간，병차능구유효지감소오검。동시，재불개변임하삼수적정황하，대우도상선전，능구명현지감소제취적차이。통과실험대비，본산법간단방편，제취각점야비상유효，상대우전통Harris산법구유경호적각점검측성능。
Harris corner detection algorithm is a classical algorithm,but it makes more serious problems and wastes more time in the large scale image. This article will put forward a new detection algorithm which can reduce the check time and lower the false corners through making non-maxima suppression within the circular region. At the same time, without changing any parameters,for image rotation,the new algorithm can significantly reduce the differences in extraction as to the image rotation. Experimental results confirm the validity of the new method and show that the new method greatly improves image corner detection performance with higher accuracy compared with the traditional algorithm.