论文部分内容阅读
通过对ＦＢＭ的Ｗｅｉｅｒｓｔｒａｓｓ－Ｍａｎｄｅｌｂｒｏｔ随机分形迈近函数的频诺分析，给出了多重分形符征的定义，并给出了判别图像的ＦＢＭ区域的方法。在ＦＢＭ区域内采用该特征进行匹配能克服复杂自然景物图像中灰度与边缘特征的不稳定性。实验表明该匹配方法能够获得比传统匹配方法，例如平均绝对差算法以及特征匹配法更高的匹配定位精度和匹配概率。
The definition of multifractal symbol is given through the frequency-norm analysis of the Weierstrass-Mandelbrot stochastic fractal function of FBM, and the method of discriminating the FBM region of the image is given. Using this feature to match in the FBM region can overcome the instability of the grayscale and edge features in complex natural scene images. Experiments show that the matching method can obtain higher matching accuracy and matching probability than traditional matching methods, such as average absolute difference algorithm and feature matching method.