基于局部分形维数的多目标检测算法

A Detection Algorithm Based on Local Fractal Dimension

在以自然景物为背景的图像中,分形维数是一种能有效地将目标从自然背景中检测出来的纹理特征。将图像的局部窗转化为一维序列的形式,在研究R/S分析法的基础上,提出了一种局部分形维数估计算法。将一维最大熵的检测结果与分形特征图的检测结果进行逻辑与运算,从而实现目标边缘的精确定位。实验表明该算法不仅可以有效地实现多目标检测同时具有较好的抗噪性能。

In the case of image with a nature scene as background, the fractal dimension feature is an effective feature that can separate the target from the nature scene. Changed the local image into the sequences of one dimension, an improved algorithm for local fractal dimention is proposed in this paper,based on the rescaled range analysis. Detection from the image of the fractal features, and combined with the threshold result, finally got the exactly verge. Experiment result shows that this method can successfully detect the targetes from the nature scene and resist the noise well.