机构地区: 华南农业大学信息学院人机交互研究中心
出 处: 《计算机应用研究》 2009年第10期3957-3959,共3页
摘 要: 提出了一种高效计算图像正交傅里叶—梅林矩的算法。该算法通过消除正交多项式中的阶乘项和提取该图像矩的公共项以提高图像矩值的计算性能。实验分析表明,与传统的直接计算方法相比,该算法可有效节省计算时间,尤其是在计算高阶连续矩情况下性能更好。 This paper proposed an efficient method to compute the orthogonal Fourier-Mellin moments for image analysis. The proposed method reduced the computation complexity by eliminating the number of factorial parts of the orthogonal polynomial and detecting the common terms in orthogonal Fourier-Mellin moments with different order and repetition. Experimental results show that the proposed method is faster than the conventional direct method, especially in case of using continuous high order moments.