【摘要】 当失配度量函数定义之后,地图匹配问题就变为了二维随机多谷函数的寻优问题。多谷快速寻优方法是作者提出的解决失配度量寻优的一种新方法。本文应用这种方法,变地图匹配中的穷举搜索为定向局部搜索,从而节省了计算,保证了匹配定位精度。本文还针对平均绝对差算法推导了门限与漏警概率的关系,给出了保证一定漏警概率的门限系数分布区间。最后,使用本文的方法在三幅航空照片上进行了大量的匹配实验,并与序贯相似性检测算法进行了比较,验证了方法的正确性,同时表明多谷快速寻优方法用于地图匹配具有定位精度高,计算量小,参数易于选择等优点。更多还原

【Abstract】 The problem of map matching becomes optimizing problem of multi-valley function after mismatching measure has been defined. Fast optimizing of multi-valley is a new method of solving the optimizing of mismatching measure proposed by authors. In this paper, appllying the method, an exhaust searching of map matching can be substituted by directional local searching, thus a lot of computation is saved and a high registration accuracy of matching is sure. Also the relationship between threshold and alarm probability is derived and a region of threshold factor which make alarm probability be less than a defined value is given for Mean Absolute Difference algorithm. Finally, after a lot of matching experiments on three different aerospace photographs using fast optimizing of multivalley and comparing with Sequential Similarity Detection Algorithm, the correctness of (his method is proved and it is shown that this method has the advantages of high registration accuracy, small computational cost, easy to select threshold et’c.更多还原