结合HIS和小波变换的自适应遥感影像融合

发布时间：2018-11-12 21:14

【摘要】：多源遥感影像融合中,通常不同地物区域对光谱特征和空间细节信息的要求有所不同。针对该特点,提出了一种IR(自适应半径搜索)显著性分析模型,实现了对多源遥感影像显著性区域和非显著性区域的识别与划分。结合HIS与小波变换融合算法,提出了改进的自适应平均梯度加权HIS融合方法,实现了对道路、农田、居民区等显著性区域的融合,更好的保留了其丰富的空间细节信息;同时提出了HIS+WT变换的融合方法,实现了对山地、林地等非显著性区域的融合,保留了较多的光谱信息。以此实现了分区域的自适应遥感影像融合。实验结果表明,本文方法能够使融合后的遥感影像既能保持较高的空间细节表现能力,又能很好地保留其光谱信息。对农业科学、森林规划以及今后森林遥感影像分类、识别等研究提供了一定的理论基础与应用价值。
[Abstract]:In multi-source remote sensing image fusion, the requirements of spectral features and spatial details are different from each other. According to this characteristic, an IR (Adaptive Radius search) significance analysis model is proposed, which can recognize and partition the salient region and non-significant region of multi-source remote sensing image. Combined with HIS and wavelet transform fusion algorithm, an improved adaptive average gradient weighted HIS fusion method is proposed to realize the fusion of significant areas such as roads, farmland, residential areas and so on. At the same time, the fusion method of HIS WT transform is put forward to realize the fusion of non-significant areas such as mountains and woodlands, and more spectral information is retained. The adaptive remote sensing image fusion is realized. The experimental results show that the proposed method can not only maintain high spatial detail representation ability, but also preserve the spectral information of the fused remote sensing image. It provides a theoretical basis and application value for agricultural science, forest planning and forest remote sensing image classification and recognition in the future.
【作者单位】：东北林业大学信息与计算机工程学院;
【基金】：黑龙江省自然科学基金资助项目(C201414) 哈尔滨市优秀学科带头人基金项目(2014RFXXJ040)
【分类号】：TP751

【相似文献】