基于松鼠搜索算法与IHS变换的遥感图像融合  

Remote Sensing Image Fusion Based on Squirrel Search Algorithm and IHS Transform

作　　者：阮理念 董燕[1] RUAN Linian;DONG Yan(Faculty of Land and Resources Engineering,Kunming University of Science and Technology,Kunming 650032,China)

机构地区：[1]昆明理工大学国土资源工程学院,云南昆明650032

出　　处：《贵州大学学报（自然科学版）》2022年第2期89-96,124,共9页Journal of Guizhou University:Natural Sciences

摘　　要：图像融合的目标是将低空间分辨率的多光谱(multispectral,MS)图像与高分辨率的全色(panchromatic,PAN)图像进行融合,得到具有高光谱和空间分辨率的图像。针对基于强度-色调-饱和度(intensity-hue-saturation,IHS)变换的融合后图像会出现光谱扭曲的问题,提出了一种结合改进的松鼠搜索算法和IHS变换的图像融合方法。首先,通过考虑融合后图像与原全色和多光谱图像之间的关系建立光谱保真和空间保真项,并引入边缘信息保持项,建立了优化目标。其次,为了克服松鼠搜索算法(squirrel search algorithm,SSA)的探索能力不足、可能导致过早收敛于局部最优的缺点,引入了自适应的捕食者存在概率策略和根据个体间距离自适应调整搜索步长的策略来平衡算法的探索和利用能力,并且引入了高斯算子和柯西算子来提高算法的局部搜索能力。最后,根据优化的控制参数,通过参数化自适应IHS变换重建目标图像。将该方法在QuickBird和IKONOS数据集上进行对比实验,结果表明方法是实用的,在保留空间信息的同时能减少光谱失真。The goal of image fusion is to fuse low spatial resolution multispectral image with high spatial resolution panchromatic image to obtain hyperspectral and spatial resolution image.Aiming at the problem of spectral distortion in fusion images based on intension-hue-saturation(IHS)transform,an image fusion method combining improved squirrel search algorithm and IHS transform was proposed.Firstly,the spectral and spatial fidelity terms were established by considering the relationship between the fused image and the original panchromatic and multispectral image,and the edge information retention term was introduced to establish the optimization objective.Secondly,in order to overcome the shortcomings of squirrel search algorithm(SSA),which may lead to premature convergence to local optimal,adaptive predator probability strategy and adaptive adjustment of search step size according to the distance between individuals were introduced to balance the algorithm’s exploration and utilization ability.Gaussian operator and Cauchy operator were introduced to improve the local search ability of the algorithm.Finally,according to the optimized control parameters,the target image was reconstructed by parameterized adaptive IHS transform.The proposed method is compared with QuickBird and IKONOS data sets,and the results show that the proposed method is practical and can reduce spectral distortion while preserving spatial information.

关 键 词：图像融合 松鼠搜索算法 IHS变换 自适应步长 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]