基于CVT-SR的合成孔径声呐高低频图像融合方法  

Fusion method for high and low frequency images of syntheticaperture sonar based on CVT-SR

作　　者：许宝华 刘传杰 王露 付五洲 XU Baohua;LIU Chuanjie;WANG Lu;FU Wuzhou(Changjiang River Estuary Bureau of Hydrological and Water Resources Survey,Hydrology Bureau of Changjiang Water Resources Commission,Shanghai 200136,China)

机构地区：[1]长江水利委员会水文局长江口水文水资源勘测局,上海200136

出　　处：《水利水电快报》2024年第10期55-61,共7页Express Water Resources & Hydropower Information

基　　金：长江水利委员会水文局科技创新基金项目(SWJ-CJX23Z12)。

摘　　要：针对合成孔径声呐高频和低频数据在浅地层中掩埋和半掩埋物探测任务中的应用需求,提出了一种结合多尺度分析和稀疏表示的合成孔径声呐高频和低频图像CVT-SR融合方法。将两张合成孔径声呐高频和低频图像进行多尺度分解,得到各自的低频系数和高频系数;按照基于稀疏表示的图像融合方法,对低频系数进行融合处理,得到低频融合系数;利用比较各层高频系数绝对值的方式对高频系数进行融合处理,得到高频融合系数。最后将高、低频融合系数进行多尺度重构,得到最终的融合图像。结果表明:相较传统方法,提出的方法融合精度更高,更清晰地呈现出了海底表面和浅地层地物信息。For the application requirements of high-frequency(HF)image and low-frequency(LF)image of synthetic aperture sonar(SAS)in sub-bottom buried and semi-buried objects detection,a CVT-SR fusion method was proposed by combining multi-scale analysis and sparse representation of SAS HF and LF images.The HF and LF images of SAS were decomposed on multi-scale to obtain their respective LF coefficients and HF coefficients.According to the image fusion method based on sparse representation,the LF coefficients were fused to obtain the LF fusion coefficient.By comparing the absolute value of HF coefficient of each layer,the HF fusion coefficient was obtained.Finally,the fusion coefficients of HF and LF were reconstructed at multi-scale to obtain the final fusion image.The results showed that the fusion accuracy of the proposed method was the highest compared with the traditional method,and the information of seafloor surface and sub-bottom objects can be presented more clearly.

关 键 词：合成孔径声呐 多尺度分析 稀疏表示 图像融合 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]