1
Access Rex and Configure Your Workspace
Begin by navigating to the Rex platform via the provided link. As a sophisticated AI-driven engine, Rex requires a structured workspace to handle high-volume visual data. Upon logging in, you will be directed to the dashboard. Here, you must initialize a new 'Analysis Project.' Unlike simple image uploaders, Rex expects geospatial context. Click 'New Project' and define the study area parameters. You will need to input the target coordinates (Latitude/Longitude) and the specific temporal window (date range) for the data you intend to analyze. This step is crucial because Rexโs computer vision algorithms rely on temporal consistency to detect changes in vegetation density and moisture levels over time. Ensure your account permissions allow for high-resolution data processing, as default limits may restrict the granularity of the terrain analysis.
Pro Tip
Pro Tip: Define your study area using a shapefile or GeoJSON polygon rather than simple bounding boxes. This allows Rex to ignore irrelevant urban noise and focus computational resources strictly on the environmental terrain you are studying, significantly improving analysis speed and accuracy.
2
Ingest High-Angle Imagery and Landscape Data
With your project configured, proceed to the 'Data Ingestion' tab. Rex specializes in parsing high-angle imagery, which typically includes satellite feeds, drone aerial photography, or LiDAR point clouds. Use the bulk upload feature to import your raw visual inputs. The system supports standard geospatial formats such as GeoTIFF, JPEG2000, and common drone export formats. As you upload, Rexโs pre-processing engine automatically aligns the imagery with its internal geospatial intelligence database. Watch the progress bar; it indicates not just file transfer, but the initial 'orthorectification' process where the AI corrects for lens distortion and terrain displacement. For best results, ensure all images share a consistent coordinate reference system (CRS), such as WGS84. If you are using disparate data sources (e.g., mixing Sentinel-2 satellite data with local drone surveys), use the 'Auto-Align' toggle to let Rex harmonize the resolutions and projections before analysis begins.
Pro Tip
Warning: Do not upload unprocessed raw drone logs without geotags. Rex requires accurate spatial metadata to map moisture and vegetation data correctly. If your imagery lacks EXIF geolocation data, use a GIS tool to embed coordinates before uploading to Rex.
3
Define Analysis Parameters: Terrain and Vegetation
Navigate to the 'Analysis Config' panel. This is the core of Rexโs utility. You must select the specific environmental metrics you wish to extract. For this tutorial, we will enable 'Terrain Characteristics,' 'Moisture Levels,' and 'Vegetation Density.' Click on 'Terrain Characteristics' to set your depth perception thresholds. Rex uses stereo-photogrammetry principles to generate digital elevation models (DEM). Adjust the 'Noise Reduction' slider if your high-angle imagery contains atmospheric haze or cloud shadows. Next, under 'Vegetation Density,' select the spectral indices you want to prioritize, such as NDVI (Normalized Difference Vegetation Index) or EVI (Enhanced Vegetation Index). Rexโs AI will automatically calculate these based on the multispectral bands present in your input. If your data is only RGB, the system will use texture analysis to estimate biomass, though multispectral data yields higher accuracy. Configure the spatial resolution of the output; for urban planning, 1-meter resolution is recommended, while regional studies may suffice with 10-meter aggregates.
Pro Tip
Tip: For urban environments with mixed greenery, enable the 'Urban Canopy Layer' filter. This prevents Rex from misclassifying lawns as forest density and helps distinguish between natural vegetation and manicured urban green spaces.
4
Run the Visual Analysis Engine
Once parameters are set, click 'Run Analysis.' This initiates Rexโs advanced computer vision pipeline. The process is not instantaneous; it involves parsing high-angle imagery, cross-referencing with historical climate data, and applying machine learning models to detect moisture anomalies. You will see a live log of processing stages: 'Geospatial Alignment,' 'Spectral Decomposition,' and 'Pattern Recognition.' During this phase, the AI identifies terrain contours and maps surface moisture retention. If you are analyzing a large area, consider using the 'Batch Mode' option, which breaks the region into tiles for parallel processing. Monitor the 'Resource Usage' widget to ensure your job does not timeout due to server load. For advanced users, you can inject custom calibration points if you have ground-truth data (e.g., soil moisture sensor readings) to fine-tune the AIโs accuracy for your specific local conditions.
Pro Tip
Common Pitfall: Do not interrupt the process during the 'Spectral Decomposition' phase. Premature cancellation results in incomplete vector data. If the job fails, use the 'Resume from Checkpoint' feature rather than restarting from scratch.
5
Interpret Visual Outputs and Heatmaps
After processing completes, switch to the 'Insights Dashboard.' Rex presents the data through interactive heatmaps and 3D terrain models. The vegetation density layer will show green gradients indicating biomass health, while moisture levels are represented by blue-to-red thermal overlays. Click on any specific region to drill down into the raw data points. Look for 'Anomaly Markers'โthese are AI-flagged areas where moisture or vegetation patterns deviate significantly from the regional average. These are critical for environmental scientists identifying drought stress or illegal land clearing. Use the 'Time Slider' to animate changes over the temporal window you defined earlier. This allows you to visualize the progression of vegetation growth or water table recession. The dashboard also provides a 'Confidence Score' for each data point; hover over low-confidence areas to see if the AI struggled with cloud cover or occlusion.
Pro Tip
Tip: Use the 'Layer Blend' tool to overlay moisture data on top of vegetation density. This composite view often reveals hidden insights, such as areas with high vegetation but low soil moisture, indicating potential irrigation inefficiencies or root-depth issues.
6
Export Data for GIS Integration
The final step is extracting the actionable insights for use in your primary GIS software (e.g., QGIS, ArcGIS). Click 'Export' and select your preferred format. For terrain data, choose 'GeoTIFF' with embedded elevation values. For vegetation and moisture, select 'GeoJSON' or 'Shapefile' to maintain vector precision. Rex also offers a 'CSV' export for tabular data, which is useful for statistical analysis in Python or R. Ensure you download the accompanying 'Metadata Report,' which details the model version, spectral indices used, and confidence intervals. This documentation is essential for peer review or regulatory compliance. Once downloaded, validate the data by importing the GeoTIFF into your GIS tool. Check that the coordinate system matches your projectโs CRS. If the data aligns correctly, you are ready to integrate Rexโs AI-derived insights into your broader environmental modeling workflows.
Pro Tip
For programmatic access, use the Rex API to automate this export. This allows you to schedule daily updates to your local GIS database without manual intervention.
7
Automate with the Rex API
For developers and data engineers, Rex provides a robust REST API to integrate visual analysis into automated pipelines. To start, generate an API Key from your Account Settings. You can then send geospatial data directly to the engine. Below is a Python example using the `requests` library to submit an analysis job and retrieve the resulting vegetation density map. This approach is ideal for monitoring large-scale environmental changes in real-time. The API accepts multipart/form-data for image uploads and JSON for configuration parameters. By automating this process, you can trigger analyses based on real-time weather alerts or satellite data feeds, ensuring your environmental models are always up-to-date with the latest terrain and moisture conditions.
```python
import requests
url = "https://api.rex.ai/v1/analyze"
headers = {"Authorization": "Bearer YOUR_API_KEY"}
data = {
"project_id": "proj_123",
"metrics": ["vegetation_density", "moisture_level"],
"resolution": "1m"
}
files = {"imagery": open("drone_survey.tif", "rb")}
response = requests.post(url, headers=headers, data=data, files=files)
print(response.json())
```
Pro Tip
Ensure your API requests include rate-limiting headers. Rex prioritizes batch jobs over individual rapid-fire requests. For high-frequency monitoring, use the WebSocket endpoint to stream results as they are processed.