BIION-D Docs

Object Detection

Run YOLO object detection on images — returns detected objects with bounding boxes.

The Object Detection module runs an object detection model on an input image. It returns detected classes, bounding boxes and confidence scores.

Object Detection module in the Action Plan editor

Configuration

Field Type Description
model_id string The ID of the YOLO model (from the Models page)
image_name string The data registry key of the input image (e.g., Camera_0.frame)
Threshold float Minimum confidence for displayed detections. Default: 0.4
is_pretrained boolean If true, always runs inference. If false, skips inference in acquisition mode

Behavior

In Inference mode (or when is_pretrained = true):

  1. Reads the input image from the data registry.
  2. Runs the model with the configured threshold.
  3. Returns a list of detected objects.

In Acquisition mode (when is_pretrained = false):

  1. Reads the input image.
  2. Skips inference — the image is saved to the dataset for training.

Inputs & Outputs

Direction Key Type Description
Input <image_name> (e.g., Camera_0.frame) image The image to analyze
Output <module_id>.detected list List of detected objects with bounding boxes and scores
Additional data stored in the registry:
Key Description
<module_id>.input Reference to the input image key
<module_id>.model_id The model used for inference

Object Detection result

Object Detection does not automatically set the pipeline compliance. Add a Plugin when detections must be converted into a custom pass/fail rule.

This module requires an image produced by a camera module in a previous block. The image_name field must reference a valid output key from an earlier block.

When Is pretrained is disabled in Acquisition mode, the image is stored for training and the detected list is empty.