论文部分内容阅读
干涉雷达配准过程中,像对间的变换模型可以通过有理多项式进行拟合,参数的解算需要均匀分布的高精度联系点,传统算法或者只重视联系点的权重,或者只考虑联系点的分布,极大地降低了多项式拟合的精度。提出一种兼顾联系点权重与分布的Voronoi图辅助配准算法,把该算法应用于中国西部某地区CosmosSkymed-1/2X波段干涉数据的处理。实验表明,相对于现有方法,该算法可以提高匹配成功率,得到更高的相关系数,获得分布更均匀的联系点,从而解算出精度更高的变换参数。
Interfering radar registration process, such as between the transformation model can be fitted by rational polynomials, parameters need to be solved uniformly distributed high-precision point of contact, the traditional algorithm or only attach importance to the weight of the point of contact, or only consider the point of contact Distribution, greatly reducing the accuracy of polynomial fitting. A Voronoi diagram-assisted registration algorithm that takes into account the weight and distribution of the contact points is proposed. The algorithm is applied to the CosmosSkymed-1 / 2X interference data processing in a certain area of western China. Experiments show that compared with the existing methods, this algorithm can improve the matching success rate, get higher correlation coefficients and obtain more uniform distribution of contact points, so as to solve the more accurate transformation parameters.