ENVI Remote Sensing Image Processing—Water Body Extraction

2. Water extraction

(1) Import the atmospherically corrected image FLAASH_result.dat.

(2) Select ToolBox——Band Ratio——Band Math in the toolbox, enter (float(b1)-float(b2))/(float(b1)+float(b2)), click Add to List, select the formula, Click OK. Note: It must be converted to float type calculation, otherwise the integer operation will get 0, as shown in Figure 2.1

Figure 2.1 Band Math tool

(3) Assign bands to b1 and b2. b1 is assigned to the Green band, and b2 is assigned to the near-infrared (NIR) band. Select the output directory and click OK to output, as shown in Figure 2.2.

Figure 2.2 Parameter setting

(4) View the results, as shown in Figure 2.3, the water body is partly white with a high value.

Figure 2.3 NDWI results

(5) The next step is to distinguish between water bodies and non-water bodies. In the toolbox, select: Classification——Decision Tree——New Decision Tree, click Node 1, and enter the calculation formula b1 GT 0.2 in expression, as shown in Figure 2.4.

Figure 2.4 New Decision Tree tool

(6) After clicking ok, a box will pop up, let select b1, click b1, select the NDWI image just generated (Figure 2.5), the effect is as follows, you can click hide at this time, at this time node1 becomes like Figure 2.6, which means : If the value is greater than 0.2, it is category 1, otherwise it is category 0; then select execute on the decision tree interface, as shown in Figure 2.7, and the result is shown in Figure 2.8.

Figure 2.5 Select b1

Figure 2.6 Decision tree Figure 2.7 Execute decision tree

Figure 2.8 Extraction results

(7) Export the water body as a vector. In the menu bar, select: classification——post classification——classification to vector, then select the binarization result just generated, then export parameter settings, select class 1, select the output path, and you can get the evf vector water body, the result As shown in Figure 2.10.

 

Figure 2.9 Parameter setting

Figure 2.10 Extract vector results

The most systematic ENVI, including land use, vegetation index, cultivated land monitoring, water quality inversion, temperature inversion, drought monitoring topics

Supongo que te gusta

Origin blog.csdn.net/hu397313168/article/details/131530968
Recomendado
Clasificación