【摘要】 选择新郑城镇区域为研究区，以高分二号(GF-2)影像为数据源，首先分别利用HSV算法、Brovey算法、PCA算法及Gram-Schmidk(GS)算法进行影像融合；然后在eCognition软件平台上，基于面向对象多尺度分割技术，利用随机森林算法对影像进行土地利用分类，并对分类结果进行精度评价。试验结果表明：不同融合算法影像融合效果明显不同，其中，GS算法融合后的影像质量最好，且分类精度最高，PCA与HSV算法次之，Brovey算法融合影像在4种融合算法中分类精度最低。更多还原

【Abstract】 This paper selects Xinzheng town area as the research area, and takes GF-2 image as the data source. Firstly, HSV algorithm, Brovey algorithm, PCA algorithm and Gram-Schmidk(GS)algorithm are used for image fusion. Then, on the eCognition software platform, based on the object-oriented multi-scale segmentation technology, the random forest algorithm is used to classify the land use of the image, and the accuracy of the classification results is evaluated. The experimental results show that the image fusion effects of different fusion algorithms are significantly different. Among them, GS algorithm has the best image quality and the highest classification accuracy, followed by PCA and HSV algorithm, and Brovey algorithm has the lowest classification accuracy among the four fusion algorithms.更多还原