Research Article
A Method of Spatial Mapping and Reclassification for High-Spatial-Resolution Remote Sensing Image Classification
Table 3
Producer and user classification accuracies for SPOT-5 image classification.
|
Class | Proposed method | Pixel-based SVM | Object-oriented SVM | SVMMV | Producer's accuracy % | User's accuracy % | Producer's accuracy % | User's accuracy % | Producer's accuracy % | User's accuracy % | Producer's accuracy % | User's accuracy % |
| Forest land |
93.00
| 88.57 | 77.00 | 88.51 | 84.00 | 83.17 | 83.00 |
96.51
| Residential land | 83.75 |
87.24
| 82.14 | 77.31 |
92.82
| 85.79 | 87.64 | 82.76 | Bare land |
64.95
| 68.48 | 50.52 | 64.47 | 44.33 |
81.13
| 46.39 | 69.23 | Industrial land | 84.00 | 93.33 | 82.00 | 91.47 |
89.00
| 91.75 | 84.00 |
95.45
| Water | 90.00 |
94.74
| 88.00 | 91.67 |
94.00
| 87.04 | 81.00 | 90.00 | Agricultural land | 80.33 |
80.83
| 80.75 | 74.14 | 80.33 | 81.00 |
86.13
| 76.61 | Road |
69.00
|
86.72
| 52.00 | 69.33 | 58.00 | 71.60 | 54.00 | 83.02 |
| Average accuracy |
80.72
|
85.70
| 73.20 | 79.56 | 77.50 | 83.07 | 74.59 | 84.80 |
| Overall accuracy |
81.32
| 74.49 | 77.67 | 77.69 |
|
|