【摘要】 在卫星遥感影像识别中，相较于海上单一环境的船舶识别，港口船舶识别由于存在集装箱、码头等大量干扰目标，显得更为困难。采用强度-色度-饱和度(Intensity-Hue-Saturation, IHS)变换、Brovey变换(Brovey Transform, BT)、ESRI全色锐化变换、简单均值变换和施密特正交变换法(Gram-Schmidt, GS)等5种融合算法，进行高分二号卫星全色和多光谱影像的融合试验，通过定性和定量评价选出适用于港口船舶影像的最优方法。结果显示GS融合方法在增加影像空间信息的同时维持了光谱保真性，其均方根误差、峰值信噪比、结构相似性等指标均优于其他4种融合方法，可用于港口船舶识别。更多还原

【Abstract】 In satellite remote sensing image recognition, compared with ship recognition in a single marine environment, port ship recognition is more difficult due to the presence of a large number of interference targets such as containers and docks. In order to improve the recognition ability of GF-2 satellite data on port ships, five fusion algorithms, i.e. Intensity Hue Saturation(IHS) Transform, Brovey Transform(BT), ESRI panchromatic sharpening Transform, simple mean Transform and Gram-Schmidt Transform(GS) were used to perform the fusion experiment of panchromatic and multispectral images, and the optimal method applicable to port ship images is selected through qualitative and quantitative evaluation. The results show that GS Transform can increase spatial information while maintaining spectral fidelity, and its mean value, root mean square error, peak signal to noise ratio, and structural similarity are superior to the other four fusion algorithms, with high recognition accuracy for port ships.更多还原