基于金字塔变换算法优化的遥感图像融合

发布时间：2018-12-17 03:56

【摘要】：为了弥补金字塔变换算法分解时数据冗余较大、融合结果不理想的缺点,提出基于金字塔变换算法优化的遥感图像融合新算法。该算法运用金字塔分解构建金字塔序列,并根据先验知识赋予相应的权重系数,通过反复迭代重建遥感图像,再利用班德文克隆选择算法优化选择,在迭代可承受的范围内,自适应地修改选择权重系数,寻求和估计合适的融合参数来优化融合效果,从而避免金字塔变换算法的经验选择。为了突出本文算法的优点,实验运用金字塔变换法、遗传算法优化金字塔变换法和粒子群算法优化金字塔变换法进行比较,从视觉效果和数理统计两个方面分析评价融合质量。实验结果表明,本文算法更符合人类的视觉感知,有利于图像的分析和信息的提取。
[Abstract]:In order to make up for the shortcomings of large data redundancy and unsatisfactory fusion results when the pyramid transform algorithm is decomposed, a new remote sensing image fusion algorithm based on pyramid transform algorithm is proposed. The algorithm uses pyramid decomposition to construct pyramid sequences, and gives corresponding weight coefficients according to prior knowledge, reconstructs remote sensing images by iterating over and over again, and then optimizes the selection by using Banderon Clone selection algorithm, which is within the range of iterative affordability. The selection weight coefficients are adaptively modified to find and estimate the appropriate fusion parameters to optimize the fusion effect, thus avoiding the empirical selection of pyramid transform algorithm. In order to highlight the advantages of this algorithm, the pyramidal transformation method, genetic algorithm optimization pyramid transformation method and particle swarm optimization pyramid transform method are used to compare and evaluate the fusion quality from two aspects: visual effect and mathematical statistics. The experimental results show that the proposed algorithm is more suitable for human visual perception and is beneficial to image analysis and information extraction.
【作者单位】：贵州师范大学喀斯特研究院;贵州省遥感中心;贵州北斗空间信息技术有限公司;
【基金】：国家自然科学基金(41661088) 贵州省科技攻关项目(黔科合GY字[2015]3001) 贵州省重大应用基础研究项目(黔科合JZ字[2014]200201)
【分类号】：TP751

【相似文献】