利用同态滤波提取高分辨率遥感影像树冠信息

Extraction of Tree-crown Information from High Spatial Resolution Remotely Sensed Imagery Based on Homomorphic Filtering

准确提取树冠信息在林业资源监测中具有重要意义。该文提出一种基于同态滤波处理的树冠信息提取方法,通过选取改进的基于高斯型高通滤波器的同态滤波函数作为传递函数,调整参数设置对高分辨率遥感影像进行同态滤波预处理;采用基于影像光谱阈值的传统分水岭算法对经同态滤波处理的高分辨率影像进行分割,并在影像光谱、形状以及纹理特征中选取6个特征构成的特征空间对树冠信息进行提取;最后对树冠信息提取结果进行定量评价。分析显示树冠信息提取总体精度为80.42%,Kappa系数为0.7939。实验结果证明所提出方法可以有效地实现树冠的信息提取,能满足部分林地树冠信息提取的精度要求,为林业管理部门监测和决策提供依据。

We proposed a new method for extracting tree-crown information from high-resolution remotely sensed imagery based on homomorphic filtering.A modified gaussian filter was selected as the homomorphic transfer function,and this filter was used to preprocess the panchromatic high resolution QuickBird satellite images by adjusting the parameters of the transfer function.In addition,the existing watershed transform based on spectral threshold method was used to segment the panchromatic image,which was processed by the homomorphic filtering.Then,the feature space consisting of spectrum,shape and texture features was selected to extract tree-crown information.Finally,240 random samples and an error matrix were applied to assess the accuracy of the extracted tree-crown information.Although confusion and error exist,this method showed promising results with an overall accuracy of 80.42% and a Kappa coefficient of 0.7939.Evaluation of the extraction results showed that the proposed method is effective in extracting tree-crown information from high spatial resolution remotely sensed imagery.