欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
传统的Harris角点检测是一种经典的角点检测算法,但不具有尺度不变性。研究了一种基于尺度空间理论的Harris角点检测方法。建立Harris的尺度空间函数表示。利用多尺度分析的思想,使在小尺度上的角点精确定位和大尺度上的角点去伪存真。从而克服了传统的Harris角点检测可能存在的角点信息丢失、角点位置偏移和易受噪声影响等问题。试验结果表明,该算法明显地提高了图像角点检测性能。
英文摘要：
Traditional Harris corner detection is a classical algorithm, but has not the scale invariant property. In this paper, an improved method of Harris corner based on the scale-space theory was described,and a scale space representation of Harris corner was established. Point corner located accurately on a small scale, eliminated the false and retained the true on a large scale by using the multi-scale idea. Then, traditional Harris detector with missing significant corners and detecting false corners due to noise was overcame. Experimental results show that the proposed rnethed is more efficient than traditional methed in detecting the corners.
同期刊论文项目
同项目期刊论文
位置：