Skip to main content

1. Imagery

This guide provides step-by-step instructions on how to deploy your imagery in TytonAI, upload imagery files, and configure imagery settings for analysis.

Overview​

The Upload Imagery module allows users to add imagery to the project either by uploading new images or selecting existing ones. These images serve as input data for AI classification processes.

Key Features​

  • Upload Raster Files: Supports uploading tiled or single .tif raster datasets.
  • Band Assignment: Specify band configurations such as Red, Green or Blue.
  • AOI Integration: Clip imagery to your Area of Interest via shapefile selection.
  • Capture Metadata: Add details like imagery name, category and date of capture.
  • Flexible File Selection: Drag and drop, browse or select from previously uploaded project files.
  • Deploy for Analysis: Finalise input settings and deploy imagery into the project.

Step-by-Step Workflow​

1. Open the Upload Imagery Panel​

  • In your project, click the "Imagery" button from the Wizard Bar.

2. Add Raster Imagery​

  • Click Select Files under Raster File(s) to open the file selection dialog.
  • You can either drag and drop your .tif files, or click the Upload button in the top-right corner to browse your local hard drive.
  • Once your files are uploaded, simply select the .tif files you want to add from your project folder to include them in your imagery layer.
  • Click Select Raster Files to confirm selection.

3. Assign Image Bands​

  • Use the dropdowns to assign each band:
    • Band 1 → Red
    • Band 2 → Green
    • Band 3 → Blue
  • Use Add Band + if additional bands like Alpha etc. are needed.

⚠️ Warning: Every tile in each image layer must have the same bands, otherwise, imagery upload will fail.

4. Enter Capture Metadata​

  • Capture Name: Give your imagery a unique, descriptive name.
  • Imagery Category: Select from:
    • UAV (0.5cm – 5cm resolution)
    • Manned Aerial (5cm – 30cm resolution)
    • Satellite (30cm - 50cm resolution)
  • Capture Date: Use the date picker to enter the capture date of your imagery.

5. Select AOI (Clip Geometry)​

  • This is required if yor imagery does not have an Alpa Band
  • Under Clip Geometry, click to select your .zip shapefile defining the Area of Interest.
  • Browse existing files or upload a new shapefile.
  • The selected shapefile will define the spatial bounds for your imagery deployment.

⚠️ Warning: Shapefiles must be uploaded to TytonAI in a compressed .zip format.
Please follow Creating a shapefile in QGIS for more info.

📌 Note: Clip Geometry Not Always Required​

If your imagery already contains an Alpha Layer (transparent background), you do not need to upload a Clip Geometry file. TytonAI will automatically use the valid data extent based on the alpha channel, eliminating the need for additional clipping.

6. Optional: Configure Additional Settings​

  • Click Show Optional Inputs to reveal advanced configuration (such as SRID, tile size, resampling).
  • These settings are optional and primarily intended for advanced workflows.

7. Deploy or Clear​

  • Click Deploy to finalise the imagery setup and load it into your project for processing.
  • Click Clear to reset all fields and start again if needed.

⚠️ Warning: When uploading multiple imagery captures to a project, ensure all datasets use the same projection (EPSG code). TytonAI supports only one projection per project—mismatched spatial references will cause errors or prevent data from being processed correctly. Always double-check your EPSG codes before uploading. Please ensure your clipped geometry is also in the same projection. follow Reprojecting a shapefile in QGIS for more info.

Selecting Existing Imagery​

  • Click Search Existing.
  • A file browser appears displaying previously uploaded image captures.
  • Use the search bar to filter results.
  • Select an image and confirm by clicking Select AOI shapefile.
Determining Band Types

Not adding the correct band types to your rasters will cause the image upload to fail.
To check the band types of your imagery:

In QGIS:

  1. Right-click the raster layer in the Layers panel.
  2. Select "Properties".
  3. Go to the "Information" tab.
  4. Look under "Bands" to see the number and types of bands.

In ArcGIS Pro:

  1. Right-click the raster layer in the Contents pane.
  2. Select "Properties".
  3. Go to the "Source" tab.
  4. Look under "Raster Information" > "Band Count".
  5. Each band's details will be listed below.

Common Combinations:

  • 3 bands = RGB
  • 4 bands = RGBA or RGBI (check metadata to confirm)
  • 5 bands = RGBIA