城市绿地信息提取中高分辨率卫星影像融合方法研究

发布时间：2018-04-09 19:03

本文选题：城市绿地　切入点：影像融合　出处：《测绘通报》2017年08期

【摘要】：利用GS变换、主成分分析、Ehlers变换、Wavelet分析、HIS变换5种方法对城区World View-2和PL-1A影像进行融合,并从影像融合质量和绿地信息提取精度两方面对融合方法的有效性进行了评价。结果表明:(1)5种融合方法中,GS变换融合的效果最好;主成分分析和Ehlers变换融合World View-2质量较好,但融合PL-1A影像质量较差;Wavelet变换、HIS变换融合两种影像质量都较差;(2)用于绿地信息提取时,GS、PCA融合影像获取的精度最高,其次为Ehlers、Wavelet融合影像,均明显高于多光谱影像的提取精度;Ehlers、Wavelet变换精度最低,绿地信息提取精度低于多光谱影像的提取精度。可以得出,影像融合可以明显地提高绿地信息提取精度,5种影像融合方法中,GS变换普适性较好,影像融合质量最好,提高分类精度效果最明显。
[Abstract]:Using GS transform, principal component analysis (PCA) and wavelet analysis (his), the World View-2 and PL-1A images in urban area were fused, and the effectiveness of the fusion method was evaluated from the aspects of image fusion quality and green space information extraction accuracy.The results show that the fusion effect of GS transform is the best among the five fusion methods, and the quality of World View-2 fusion by principal component analysis and Ehlers transform is better.However, the quality of fused PL-1A images is poor. Wavelet transform and his transform fusion are both of poor quality. In the case of green space information extraction, the accuracy of PL-1A fusion image is the highest, and the second is EhlersWare fusion image, the second is EhlersWare fusion image, and the second is Ehlerswavelet fusion image.The accuracy of Ehlersberg wavelet transform is the lowest, and the precision of green space information extraction is lower than that of multispectral image.It can be concluded that image fusion can obviously improve the accuracy of green space information extraction. Among the five image fusion methods, the GS transform is generally applicable, the quality of image fusion is the best, and the effect of improving classification accuracy is the most obvious.
【作者单位】：淮海工学院测绘信息与海洋学院;
【基金】：国家自然科学基金(31070626;31270745) 江苏省第九批次“六大人才高峰”项目江苏省“333”工程人才项目
【分类号】：P237;TU985

【相似文献】