GeoAI Semantic Classification

Apply a trained semantic-segmentation model to a high-resolution raster and produce a pixel-level classified raster.

Remote Sensing CoreEnd-User DocumentationTheory + Formula + Parameters

1. What This Feature Does

Apply a trained semantic-segmentation model to a high-resolution raster and produce a pixel-level classified raster.

Cloud workflow: choose data → configure the scientific method → run → review the resulting layer/report. No programming is required.

2. Recommended Workflow

Choose a trained semantic model from the project. ↓ Choose the raster and RGB mapping if needed. ↓ Set inference tile size and compute device. ↓ Run tiled inference. ↓ Review class boundaries and validate against independent labels.

3. Theory, Methods & Equations

Semantic Segmentation Inference

The model produces a score/logit for every class at every pixel; softmax converts scores into class probabilities and the highest probability becomes the output class.

p(c|x)=exp(z_c)/Σ_jexp(z_j)

ĉ=argmax_c p(c|x)

Best used when: a custom semantic model has already been trained for the required classes.

4. Input Data

Input	Type	Requirement	Description
Raster Input	raster	Required	High-resolution raster to classify.

5. Parameters Available in the Application

Parameter	Default	Choices / Range	How to Use It
Model Source	custom	Custom — My Project Models
Trained Semantic Model	—	—	Semantic models generated by GeoAI Model Training and stored in this project's files catalog. Shown for: custom
RGB Bands Override	—	—	Leave empty to use the RGB mapping stored in the checkpoint.
Inference Tile Size	512	128 to 2048
Device	auto	Auto, CPU, CUDA GPU, Apple MPS
Output Nodata	255	0 to 65535

6. Output & Interpretation

The result should be interpreted according to the selected method and the physical meaning of the input data. Preserve source units, coordinate reference information, NoData meaning, acquisition date, and preprocessing level when comparing results.

Scientific interpretation: an algorithm can produce a numerically valid result even when the input data are unsuitable. Always check masks, units, sensor characteristics, spatial resolution and reference data.

Important: Inference tile size affects memory and contextual coverage. A larger tile can provide more context but requires more GPU/CPU memory.

7. Best Practices

Use analysis-ready inputs and remove invalid/cloud/noise artifacts that are not part of the target phenomenon.
Choose parameters from the physical scale of the data, not only from visual appearance.
Keep categorical and continuous rasters conceptually separate when selecting interpolation/resampling methods.
Compare the result with the source image and independent reference information.
Document the settings used when results will be compared across dates, sensors or study areas.

← GeoAI Object Extraction GeoAI Model Training →