3D Target Object Recognition

You are currently viewing the documentation for the latest version (2.2.0). To access a different version, click the "Switch version" button located in the upper-right corner of the page.

■ If you are not sure which version of the product you are currently using, please feel free to contact Mech-Mind Technical Support.

Function

This Step is integrated with vision processing functions such as point cloud preprocessing, deep learning, and removal of overlapped objects, facilitating a rapid recognition of target objects.

Usage Scenario

This Step is suitable for workpiece loading scenarios. It is capable of recognizing workpieces of various shapes and stacking methods, including separate arrangements, orderly single-layer stacking, orderly multi-layer stacking, and random stacking. This Step usually follows the Capture Images from Camera Step and is followed by the pose adjustment Step.

Input and Output

Input

Input port Data type Description

Camera Depth Map

Image/Depth

Original depth map of the object.

Camera Color Image

Image/Color

Original color image of the object.

Output

Output port Data type Description

Pick Points

Pose[]

List of pick points of the target objects.

Pick Point Info

JsonValue

The names of the pick points and their corresponding target objects.

Parameter Description

Parameter Description

Select Target Object

Parameter description: Select a target object to recognize from the drop-down list. The target object should be added in advance via the config wizard.
Tuning instructions: Click Config wizard or the Config wizard button on the Step block to enter the 3D Target Object Recognition Tool to configure the model. For detailed instructions, refer to 3D Target Object Recognition Tool.

After selecting Trigger Control Flow Given No Output for the “3D Target Object Recognition” Step, the control flow will still be triggered even if the output of the Step is empty. Note:

  • Once “Trigger Control Flow Given No Output” was selected, the “3D Target Object Recognition” Step could still output “Raw Point Cloud Captured by Camera” and “Preprocessed Point Cloud”.

  • The “Trigger Control Flow Given Output” feature will always take effect.

Is this page helpful?

You can give a feedback in any of the following ways:

We Value Your Privacy

We use cookies to provide you with the best possible experience on our website. By continuing to use the site, you acknowledge that you agree to the use of cookies. If you decline, a single cookie will be used to ensure you're not tracked or remembered when you visit this website.