DeltaXStudio — Vision

On this page Choose your detection method After detection works

DeltaXStudio — Vision

Choose one detection method below. Each guide contains the complete workflow for that choice, from image preparation to a live detection check.

Complete Calibration first. Keep the camera position, crop, focus, and lighting consistent with operation, and leave robot movement stopped during detection tests.

Choose your detection method

AI

Recognize object categories from a trained model. Capture and download images in DeltaXStudio, upload them to DeltaXHub AI Hub for labeling and training, then download RKNN and activate it in DeltaXStudio.

Model trainingObject classes

Capture images, train, and deploy →

Color filtering

Detect objects whose color stands out from the background. Build an HSV and contour pipeline in Blockly.

No trainingBlockly

Open guide →

AI + Template Matching

Use the shared AI setup, then add saved reference templates to refine or filter its detections.

Model trainingReference images

Open guide →

Color + Template Matching

Find objects by color, then use reference images to refine or reject candidates before tracking.

No trainingBlockly + reference images

Open guide →

Template preparation is included in both combined guides. Choose the complete workflow you intend to use.

After detection works

Check one detection per object, the class index, center, and orientation. Then continue to Pick & Place to configure destinations and test one object.

Found something that could be clearer?Suggest an edit

What are you working on?

Try a command like G28, a model, or a topic like wiring.