| ||||||||||||||||
| ||||||||||||||||
| ||||||||||||||||
This module has no screenshots and no videos
| ||||||||||||||||
Module DocumentationThis module runs a deep neural network using the OpenCV DNN library. Classification networks try to identify the whole object or scene in the field of view, and return the top scoring object classes. Detection networks analyze a scene and produce a number of bounding boxes around detected objects, together with identity labels and confidence scores for each detected box. Semantic segmentation networks create a pixel-by-pixel mask which assigns a class label to every location in the camera view. To select a network, see parameter Serial messagesFor classification networks, when object classes are found with confidence scores above DO category:score category:score ... category:score where category is a category name (from See Standardized serial messages formatting for more on standardized serial messages, and Helper functions to convert coordinates from camera resolution to standardized for more info on standardized coordinates. For object detection networks, when detections are found which are above threshold, one message will be sent for each detected object (i.e., for each box that gets drawn when USB outputs are used), using a standardized 2D message:
See Standardized serial messages formatting for more on standardized serial messages, and Helper functions to convert coordinates from camera resolution to standardized for more info on standardized coordinates. | ||||||||||||||||
| ||||||||||||||||
|