FusionOfClassifications¶
在分类标签的基础上融合同一图像的多个分类地图。
描述¶
这个应用程序允许您融合多个分类映射，并生成单个更健壮的分类映射。融合可以通过多数投票的方式完成，也可以通过类标签上的Dempster Shafer组合方法来完成。
- 多数投票：对于每个像素，选择得票数最高的类别。
- Dempster Shafer：对于每个像素，选择信任函数最大的类别标签。该置信度函数通过大量置信度的Dempster Shafer组合来计算，并指示每个输入分类映射为每个标签值表示的置信度。此外，信任量是基于每个分类图的输入混淆矩阵，通过使用准确率或召回率，或总体准确率，或Kappa系数。因此，对于Dempster Shafer融合，每个输入分类映射需要与其对应的输入混淆矩阵文件相关联。
- 在分类地图的融合中，不处理带有NODATA标签的输入像素。此外，所有输入分类器都被设置为NODATA的像素在输出融合图像中保持此值。
- 在票数相等的情况下，未决定的标签被归因于像素。
参数¶
Input classifications -il image1 image2... Mandatory
List of input classification maps to fuse. Labels in each classification image must represent the same class.
The output classification image -out image [dtype] Mandatory
The output classification image resulting from the fusion of the input classification images.
Fusion method -method [majorityvoting|dempstershafer] Default value: majorityvoting
Selection of the fusion method and its parameters.
- Majority Voting
Fusion of classification maps by majority voting for each output pixel. - Dempster Shafer combination
Fusion of classification maps by the Dempster Shafer combination method for each output pixel.
Dempster Shafer组合选项¶
Confusion Matrices -method.dempstershafer.cmfl filename1 filename2... Mandatory
A list of confusion matrix files (.csv format) to define the masses of belief and the class labels. Each file should be formatted the following way: the first line, beginning with a '#' symbol, should be a list of the class labels present in the corresponding input classification image, organized in the same order as the confusion matrix rows/columns.
Mass of belief measurement -method.dempstershafer.mob [precision|recall|accuracy|kappa] Default value: precision
Type of confusion matrix measurement used to compute the masses of belief of each classifier.
- Precision
Masses of belief = Precision rates of each classifier (one rate per class label). - Recall
Masses of belief = Recall rates of each classifier (one rate per class label). - Overall Accuracy
Mass of belief = Overall Accuracy of each classifier (one unique value for all the class labels). - Kappa
Mass of belief = Kappa coefficient of each classifier (one unique value for all the class labels).
Label for the NoData class -nodatalabel int Default value: 0
Label for the NoData class. Such input pixels keep their NoData label in the output image and are not handled in the fusion process. By default, 'nodatalabel = 0'.
Label for the Undecided class -undecidedlabel int Default value: 0
Label for the Undecided class. Pixels with more than 1 fused class are marked as Undecided. Please note that the Undecided value must be different from existing labels in the input classifications. By default, 'undecidedlabel = 0'.
实例¶
从命令行执行以下操作：
otbcli_FusionOfClassifications -il classification1.tif classification2.tif classification3.tif -method dempstershafer -method.dempstershafer.cmfl classification1.csv classification2.csv classification3.csv -method.dempstershafer.mob precision -nodatalabel 0 -undecidedlabel 10 -out classification_fused.tif
来自Python的评论：
import otbApplication
app = otbApplication.Registry.CreateApplication("FusionOfClassifications")
app.SetParameterStringList("il", ['classification1.tif', 'classification2.tif', 'classification3.tif'])
app.SetParameterString("method","dempstershafer")
app.SetParameterString("method.dempstershafer.mob","precision")
app.SetParameterInt("nodatalabel", 0)
app.SetParameterInt("undecidedlabel", 10)
app.SetParameterString("out", "classification_fused.tif")
app.ExecuteAndWriteOutput()