Instructions for use
Perform classification of ground objects on the imagery, i.e., analyze using the spectral information of various ground objects in remote sensing images and geo-information, assign semantic category labels to each pixel with semantic information in the image to achieve classification of ground objects, including: buildings, forest land, grassland, paddy fields, agricultural land, etc. Analogous to binary classification, classification of ground objects uses a neural network model to determine which class each pixel in the image belongs to among the classes of interest, used for multi-class ground object judgment, and calculates information such as the location, boundaries, and area of the classes of interest through pixels.
Parameter description
-
Source Dataset：Select the imagery that requires classification of ground objects, supporting two types: dataset and folder.
-
-
If file type selects Dataset, it supports input of file-based image data (such as *.tif, *.img, etc.), image dataset, and mosaic dataset.
-
If file type selects Folder, you need to input the image folder path.
-
Notes:- Only supports 8-bit unsigned format imagery.
- Supports three-band and multi-band imagery (imagery with 4 or more bands), and the band order of the imagery must be consistent with the sample image used during model training.
-
Model File：Select Model File (*.sdm)。
-
Tile Overlap (Pixels)：During the classification of ground objects process, due to the structural characteristics of convolutional neural networks, data at the edges of tiles may not be fully predicted. To improve prediction quality, tiles need to be overlapped. This parameter is in pixels and can be set to one-eighth to one-quarter of the tile size. The larger the tile overlap pixels, the longer the time required to infer the entire image. The tile size corresponding to the model is recorded in the tile_size parameter of the model file (*.sdm). For example, if the tile size is 1024, the tile overlap pixels can be set to 128-256.
-
Batch Size：Due to the large volume of remote sensing image data, and limitations of computer performance, the model cannot be read in at once. Therefore, during inference, a block-based reading and processing method is used. Batch size refers to the number of slices inferred simultaneously. Appropriately increasing the batch size can improve interpretation efficiency, but it is limited by the memory of the inference device (GPU memory for GPU inference or RAM for CPU inference).
-
Processor Type：You can use the computer's Central Processing Unit (CPU) or Graphics Processing Unit (GPU) to process data.
-
GPU ID：Specify the GPU ID for data processing. Default is 0. If "Processor Type" selects CPU, this value is fixed to -1. If using GPU, you need to query the GPU ID through the "nvidia-smi" command in the system command line. The GPU ID is based on the query result. If there are multiple GPUs, you can specify the GPU identifier for data processing. If multiple GPUs need to be used, separate them with English commas ",", such as "0,2,3", indicating the use of GPUs with IDs 0, 2, and 3.
-
Other Parameters：Check Other Parameters, you can input a region dataset with the same coordinate system as the source data to limit the inference range.
-
Result Data：The result of classification of ground objects can be output as a vector dataset or raster file.
-
If selecting Vector Dataset, you need to specify the data source for saving the vector dataset and the vector dataset name. If it is batch inference, the original image name will be appended with "_" to the vector dataset name to distinguish multiple classification results of multiple images.
-
If selecting Raster File, you need to specify the folder path for saving the raster file and the raster file name. If it is batch inference, the original image name will be appended with "_" to the raster file name to distinguish multiple classification results of multiple images.
-
Related Topics