Figures & data
Figure 1. (a) Location of the Zoige wetland in China, (b) the boundary of the study area, (c) the true color composite image of the Landsat 8 OLI data of the whole study area
![Figure 1. (a) Location of the Zoige wetland in China, (b) the boundary of the study area, (c) the true color composite image of the Landsat 8 OLI data of the whole study area](/cms/asset/db1dda9c-111a-457c-a1c3-78869f82a85a/tgrs_a_1932126_f0001_oc.jpg)
Table 1. List of input features used for wetland cover classification. NDVI = normalized difference vegetation index, NDWI = normalized difference water index, MEAN = mean, HOMO = homogeneity, ASM = angular second moment, DIS = dissimilarity, ENTR = entropy, VAR = variance
Figure 2. Locations of reference samples in the study area and typical examples of wetland covers in the Landsat 8 OLI images. AM = alpine meadow, MM = marsh meadow, M = marsh, RL = river and lake, FP = floodplain, S = sediments, RA = residential area, BR = bedrock
![Figure 2. Locations of reference samples in the study area and typical examples of wetland covers in the Landsat 8 OLI images. AM = alpine meadow, MM = marsh meadow, M = marsh, RL = river and lake, FP = floodplain, S = sediments, RA = residential area, BR = bedrock](/cms/asset/064427d6-f0f8-459d-911f-e7e186979a37/tgrs_a_1932126_f0002_oc.jpg)
Table 2. Descriptions of wetland cover categories used in this study
Figure 3. The flowchart of wetland cover classification. ANN = artificial neural network, MBANN = MultiBoost artificial neural network, RANN = rotation artificial neural network, VGG = visual geometry group, RF = random forests
![Figure 3. The flowchart of wetland cover classification. ANN = artificial neural network, MBANN = MultiBoost artificial neural network, RANN = rotation artificial neural network, VGG = visual geometry group, RF = random forests](/cms/asset/f410302b-c6f3-495f-ae5c-a4d269b7b067/tgrs_a_1932126_f0003_b.gif)
Figure 4. Illustration of the MBANN method. ANN = artificial neural network, MBANN = MultiBoost artificial neural network, AB = AdaBoost
![Figure 4. Illustration of the MBANN method. ANN = artificial neural network, MBANN = MultiBoost artificial neural network, AB = AdaBoost](/cms/asset/9c75540e-cedc-4bfb-9692-f893cf99211f/tgrs_a_1932126_f0004_oc.jpg)
Figure 5. Illustration of the RANN method. ANN = artificial neural network, RANN = rotation artificial neural network, PCA = principal component analysis
![Figure 5. Illustration of the RANN method. ANN = artificial neural network, RANN = rotation artificial neural network, PCA = principal component analysis](/cms/asset/437b0f2f-0bb1-43d8-99af-bb24317d092f/tgrs_a_1932126_f0005_oc.jpg)
Table 3. Classification accuracy evaluation and comparison of various classification methods. ANN = artificial neural network, MBANN = MultiBoost artificial neural network, RANN = rotation artificial neural network, VGG11 = visual geometry group, RF = random forests. AM = alpine meadow, MM = marsh meadow, M = marsh, RL = river and lake, FP = floodplain, S = sediments, RA = residential area, BR = bedrock
Table 4. Z-score test with associated probability value (P) for model pairs. ANN = artificial neural network, MBANN = MultiBoost artificial neural network, RANN = rotation artificial neural network, VGG11 = visual geometry group, RF = random forests
Figure 6. Wetland cover maps obtained by the five classifiers: (a) artificial neural network, (b) MultiBoost artificial neural network, (c) rotation artificial neural network, (d) visual geometry group, and (e) random forests
![Figure 6. Wetland cover maps obtained by the five classifiers: (a) artificial neural network, (b) MultiBoost artificial neural network, (c) rotation artificial neural network, (d) visual geometry group, and (e) random forests](/cms/asset/351c603f-477b-4feb-98df-e5f9b9ca3872/tgrs_a_1932126_f0006_oc.jpg)
Figure 7. Area percentages of wetland cover classes predicted by different classification methods. ANN = artificial neural network, MBANN = MultiBoost artificial neural network, RANN = rotation artificial neural network, VGG11 = visual geometry group, RF = random forests
![Figure 7. Area percentages of wetland cover classes predicted by different classification methods. ANN = artificial neural network, MBANN = MultiBoost artificial neural network, RANN = rotation artificial neural network, VGG11 = visual geometry group, RF = random forests](/cms/asset/3a60f918-b421-4cb2-bc72-61a6a145684f/tgrs_a_1932126_f0007_oc.jpg)
Figure 8. Robustness evaluation concerning the impact of the data size on the classification accuracy of classifiers. RANN = rotation artificial neural network, MBANN = MultiBoost artificial neural network, ANN = artificial neural network, RF = random forests, VGG11 = visual geometry group
![Figure 8. Robustness evaluation concerning the impact of the data size on the classification accuracy of classifiers. RANN = rotation artificial neural network, MBANN = MultiBoost artificial neural network, ANN = artificial neural network, RF = random forests, VGG11 = visual geometry group](/cms/asset/829031f6-249e-485b-ace6-fc7b18ca6983/tgrs_a_1932126_f0008_oc.jpg)