geointel/docs/UI_PAGE_SPECIFICATIONS.md

GeoIntel Kempen — UI Page Specifications v1.0

The UI must feel like a modern GeoAI workbench, not a generic admin dashboard.

Global layout

• Left navigation: product sections.
• Center: map/workbench content.
• Right panel: context, analysis settings, result details.
• Bottom or top status area: job status and notifications.
• Dark professional theme by default, with enough contrast.

Main pages

1. Home

Purpose

Start or resume work.

Components

• Product intro: GeoAI Workbench for the Kempen.
• New project button.
• Open demo scenario button.
• Recent projects.
• System readiness cards: backend, database, storage, AI models.

2. Project Workspace

Purpose

Project overview.

Components

• Project summary.
• Areas list.
• Datasets list.
• Recent analysis runs.
• Quick actions: upload raster, upload vector, draw area, run detection.

3. Map Workbench

Purpose

Spatial exploration and layer inspection.

Components

• MapLibre map.
• Layer manager.
• Draw controls.
• Feature inspector.
• Opacity controls.
• CRS/bounds information.

Required actions

• Draw area.
• Save area.
• Toggle layers.
• Click feature for attributes.

4. Dataset Manager

Purpose

Upload, inspect and manage raster/vector datasets.

Components

• Upload zone.
• Dataset table.
• Metadata detail panel.
• Processing status.
• Preview button.

API calls

• POST /projects/{id}/datasets/upload
• GET /projects/{id}/datasets
• GET /datasets/{id}/metadata

5. Raster Lab

Purpose

Inspect and process rasters.

Components

• Raster selector.
• Metadata cards.
• Band selector.
• Histogram.
• Preview.
• Clip controls.
• Tile generation controls.

6. Vector Lab

Purpose

Inspect and process vector layers.

Components

• Vector selector.
• Attribute schema.
• Geometry summary.
• Operation buttons: clip, buffer, intersect, difference, dissolve.
• Result layer list.

7. Detection Lab

Purpose

Run object detection on raster imagery.

Components

• Raster selector.
• Model selector.
• Class selector.
• Confidence threshold.
• Tile size/overlap.
• Run button.
• Detection results table.
• Map overlay.

8. Segmentation Lab

Purpose

Create masks and polygons from imagery.

Components

• Raster selector.
• Model selector.
• Mode selector.
• Threshold controls.
• Mask preview.
• Polygonized output table.

9. QA/QC Lab

Purpose

Validate predictions against reference layers such as GRB.

Components

• Prediction layer selector.
• Reference layer selector.
• Class filter.
• IoU threshold slider.
• Metrics cards.
• TP/FP/FN map overlays.
• Findings table.

10. Change Detection Lab

Purpose

Compare timepoints or analysis outputs.

Components

• Dataset/layer A selector.
• Dataset/layer B selector.
• Method selector.
• Threshold controls.
• Added/removed/modified layers.
• Summary cards.

11. Exports

Purpose

Download results.

Supported exports

• GeoJSON
• CSV
• COCO later
• YOLO later
• masks
• report

12. Settings

Purpose

System configuration.

Components

• Data source settings.
• Storage paths.
• AI model paths.
• Processing defaults.
• Cache management.

UX rules

• Every empty state must explain what the user should do next.
• Every long operation must create a visible job.
• Every analysis result must have export options.
• Every map layer must be inspectable.
• Do not create fake results to make pages look full.