| 基于数学形态学和最大似然法的遥感图像分类研究 |
| |
| 摘 要： | 随着高空间分辨率遥感影像应用范围的不断扩大,传统基于灰度值的遥感图像分类方法很难满足实际需要.该文通过数学形态学方法,对高空间分辨率遥感全色图像进行处理,通过交互式选择训练区,构造包含形态学梯度、高帽变换和灰度均值的三维特征向量,利用Bayes最大似然分类器对高空间分辨率遥感图像不同土地利用类型进行自动识别,改善了分类精度.这种分类方法,可以用于指导森林资源监测、土地利用现状调查和国土荒漠化监测与评价的工程实践.
|
| 关 键 词： | 数学形态学 最大似然法 高空间分辨率遥感影像 图像分割 |
Automatic segmentation of high-resolution remote sensing image based on mathematical morphology and maximum likelihood estimate. |
| |
| Abstract: | Presently conventional methods hardly kept pace with the wide application of high-resolution remote sensing image.In this paper,a new classifying method based on mathematical morphology and maximum likelihood estimate was used successfully to segment high-resolution remote sensing image and automatically classify the usage of land.By means of interactively selecting samples,an eigenvector with three elements such as morphological gradient,top-hat transformation and grayscale mean was constructed.Then RS image segmentation was fulfilled with the maximum likelihood classification.Experiment processing was also listed in the passage.The classified image clearly showed that the method was efficient and significant,and it could be put into practice such as forestry resource assessment, land usage investigation and desertification monitoring and estimate. |
| |
| Keywords: | mathematical morphology maximum likelihood estimate high-resolution remote sensing image image segmentation |
|
|
