摘要
Harris角点检测是一种经典的角点检测算法,现实中应用比较广泛,但不具有尺度变化特性,所以在图像的角点提取中往往改变参数的选择也得不到满意的提取效果。为了改变其单一尺度的特性,使得角点提取更加精确和有效,文中将多尺度空间和模糊系数引入到该算法中,在多个尺度下结合Harris算法对角点进行提取。该算法融合了多个尺度的特征信息,克服了单一尺度的Harris角点检测可能存在的角点信息丢失和易提取伪角点等问题。通过对比实验,文中算法明显地提高了图像角点检测性能。
Harris comer detection is a classical algorithm and used wildly now,but has not the property of scale invariant. So it often does not get a satisfied result in the image of the comer detection. For modifying its single - scale and making cerner detection more accurate and valid, in this paper, multi- scale - space and fuzzy parameter are introduced into the Harris algorihm. Harris multi- scale comer detection is used to comer extraction by multi - scale- space. It not only includes the feature information at several scales,but also overeomes the drawback that the single- scale Harris detector usually leads to either missing significant comers or detecting false comers. Compared with Hanrris algorithm,the presented algorithm is more efficient in detecting the comers with accurate lccation.
出处
《计算机技术与发展》
2010年第4期58-60,64,共4页
Computer Technology and Development
基金
兵器预研支撑基金(62301110113)
