【摘要】 文章通过GF-2卫星全色及多光谱数据融合试验,比较了常用的4种影像融合方法,从主观定性评价和客观定量评价两方面分析融合后的影像质量,最后采用最大似然法对融合影像分类比较,筛选较适合GF-2影像的融合方法。试验结果表明:四种融合方法中,Pansharp变换取得了较好的融合效果,在主观定性和客观定量评价中效果最优,在分类试验中总体分类精度最高。总体而言,Pansharp变换优于其他融合算法,适用于GF-2卫星在林业行业的应用。研究力图为林业中影像大规模的融合应用提供参考。更多还原

【Abstract】 This paper compared 4 kinds of image fusion methods through the GF-2 panchromatic and multispectral data fusion test. The subjective qualitative evaluation and objective quantitative evaluation were applied to analyze the quality of the fusion images, which were classified by the maximum likelihood method to select the suitable fusion method for GF-2 image. Test results showed that, among the four fusion methods, Pansharp transform achieved better fusion effect. It had the best subjective and objective qualitative evaluation; It had the highest overall classification accuracy in the maximum likelihood classification test. The experiment results indicated that the Pansharp transform had better fusion effect than other fusion algorithms, and it was suitable for the application of GF-2 in forestry industry. The research aimed to provide reference for the application of image fusion in the forestry sector.更多还原