改进的Hausdorff距离和遗传算法在图像匹配中的应用

研究模板和图像间的有效匹配，将部分Ｈａｕｓｄｏｒｆｆ距离的计算进行改进，提出一种改进的部分Ｈａｕｓｄｏｒｆｆ距离作为检测模板和图像中物体轮廓相似性的测试，可以较大地减少计算量，同时把遗传算法引入图像匹配识别，由于遗传算法的高并行性和鲁棒性，可以较快地完成全局搜索，而不会陷入局部最优，因此该算法和改进的Ｈａｕｓｄｏｒｆｆ距离相结合能有效地检测出具有平移、旋转和尺度变化的物体，该方法可以应用于实际图像识别和匹配中。

The Application of Improved Hausdorff Distance and Genetic Algorithm in Image Matching

Effectively matching the model with the image is studied. The directed Hausdorff distance is improved to measure the degree of similarity between models and images, which can reduce the computational complexity. Genetic algorithm is used to search the interested objects. Because the genetic algorithm is a parallel and robust algorithm, the combination of genetic algorithm and improved Hausdorff distance can be used to find the global optimum results. The experimental results show that the proposed method can efficiently detect the objects that are changed in translation, rotation and scale. The method can be used in image identification and matching in practice.