Binary Classification
Feature Description
Binary classification utilizes spectral information and geo of ground objects in remote sensing images to classify specific elements through deep learning algorithms. It determines whether image pixels belong to the target category, generating binary raster data with values 0 (non-target) and 1 (target).
This function performs pixel-level fine-grained interpretation with flexible raster results. The classification outcomes can be optimized through post-processing steps, and converted into vector data for further analysis.
Binary classification is typically used for extracting distinct single categories like roads, rivers, and buildings. It enables calculation of target category positions, boundaries, and areas through pixel analysis.
Parameter Description
| Parameter Name | Default Value | Description | Type |
|---|---|---|---|
| File Type | Dataset | Specifies the file type, which can be dataset or folder type. | Boolean |
| Model File | Specifies the model file (*.sdm) for binary classification | String | |
| Tile Overlap (Pixels) | 0 | Specifies the tile overlap in pixels. Reduces prediction inaccuracies at tile edges. Larger overlaps increase processing time | Integer |
| Batch Size | 1 | Specifies the number of images processed simultaneously during inference | Integer |
| Processor Type | GPU | Specifies the processor type | String |
| GPU ID(s) | 0 | Input GPU IDs for multi-card inference. Example: "0,1,2" will be converted to "012" | String |
| Other Parameter Settings (Optional) |
false | Check to enable bounds dataset/datasource parameter configuration | Boolean |
| Result Datasource | Specifies the datasource storing the result dataset | Datasource | |
| Result Dataset | binclassify_result | Specifies the name of the output dataset | String |
Output
| Parameter Name | Description | Type |
|---|---|---|
| Result Dataset | Result dataset | DatasetVector |