Use the Deep Learning Labeling Tool to create labeled data. The Deep Learning Labeling Tool simplifies the process of drawing rectangle annotations for object detection, or ROIs for pixel segmentation. This is the preferred method for creating object detection rasters or label rasters (collectively called labeled data) for training.

  1. Choose one of the following options to start the Labeling Tool:
    • In the ENVI Toolbox, select Deep Learning > Deep Learning Labeling Tool.
    • In the Deep Learning Guide Map, click one of the following button sequences:
      • Pixel Segmentation > Train a New Pixel Model > Label Rasters.
      • Object Detection > Train a New Object Model > Label Rasters.
      • Grid > Train a New Grid Model > Label Rasters.

  2. Select File > New Project from the Labeling Tool menu bar. The Create New Labeling Project dialog appears.

  3. From the Project Type drop-down list, select one of the following:

    • Pixel Segmentation
    • Object Detection. If you accessed the Labeling Tool from the Guide Map, this option is automatically selected and you cannot change it.
  4. Enter a Project Name.
  5. Click the Browse button next to Project Folder and select an empty folder in which to store project files.
  6. For Enhance Display, select Yes if you want to apply an additional small stretch to the processed data to suppress noise and enhance feature visibility. The optional stretch is effective for improving visual clarity in imagery acquired from aerial platforms or sensors with higher noise profiles. The option you choose cannot be changed after the project has been created.

  7. For Visual RGB, select Yes if you want to encode the output raster as a three-band RGB composite (red, green, blue) for color image processing. This ensures consistent band selection from ENVI display types (such as RGB, CIR, and pan) and supports integration of diverse data sources (such as MSI, panchromatic, and VNIR) without band mismatch. The option you choose cannot be changed after the project has been created.

  8. Click OK.

When you create a new project and define your training rasters, a subfolder will be created under the project folder for each training raster you selected as input. Each subfolder contains the annotations used to label features.

If you selected Yes for the Enhanced Display and/or Visual RGB parameters, a source_rasters subfolder will be created under the project folder. The source_rasters subfolder contains the stretched and/or encoded version of the original rasters, leaving the original source files unaltered and in their original location. When Enhanced Display and/or Visual RGB is set to No, the source files remain unaltered in their original location and no additional source file directory is created.

A file named source_raster.json is also saved in each training raster subfolder. This is a simplified version of the training raster, where all of its information is condensed into JSON code. If you move the project folder to a different location, the source_raster.json file in each project subfolder tells ENVI Deep Learning where to find the training raster. This way, you do not have to keep track of file locations yourself.

To restore a previously created project, select File > Open Project in the Labeling Tool menu bar. Navigate to your project folder and select the file deep_learning_labeling.json.

After setting up a project, the next step is to define your output classes. Refer to the following topics for additional steps, depending on the option you choose:

See Also

Label Features Background