Object-Bin Segmentation

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 segments the target objects and the bin from the input depth map and color image based on the Object-Bin Segmentation Model Package. It outputs the target object mask and the bin mask, and provides visualization results.

Usage Scenario

This Step is suitable for scenarios where target objects need to be effectively separated from the bin. It typically follows the image acquisition Steps and precedes the point cloud extraction Steps.

Go to Download Center to get the Object-Bin Segmentation deep learning model package.

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.

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
Visualization Output	Image/Color	Visualized results.
Object Presence	Bool	Detection result indicating whether target objects are present in the input image. true indicates that target objects exist, and false indicates that no target objects exist.
Object Mask Image	Image/Color/Mask	Mask image of segmented target objects.
Bin Mask Image	Image/Color/Mask	Mask image of the segmented bin.

Output port

Data type

Description

Visualization Output

Image/Color

Visualized results.

Object Presence

Bool

Detection result indicating whether target objects are present in the input image. true indicates that target objects exist, and false indicates that no target objects exist.

Object Mask Image

Image/Color/Mask

Mask image of segmented target objects.

Bin Mask Image

Image/Color/Mask

Mask image of the segmented bin.

System Requirements

The following system requirements need to be met when using this Step.

CPU: needs to support the AVX2 instruction set and meets any of the following conditions:
- IPC or PC without any discrete graphics card: Intel i5-12400 or higher.
- IPC or PC with a discrete graphics card: Intel i7-6700 or higher, with the graphics card not lower than GeForce GTX 1660.
This Step has been thoroughly tested on Intel CPUs but has not been tested on AMD CPUs yet. Therefore, Intel CPUs are recommended.
GPU: GeForce GTX 1660 or above (if with a discrete graphics card).

Parameter Description

Model Package Settings

Parameter	Description
Model Manager Tool	Parameter description: This parameter is used to open the deep learning model package management tool and import the deep learning model package. The model package file is a “.dlkpack” file exported by Mech-DLK. Tuning instruction: Please refer to Deep Learning Model Package Management Tool for the usage.
Model Name	Parameter description: This parameter is used to select the model package that has been imported for this Step. Tuning instruction: After importing a deep learning model package with the Deep Learning Model Package Management Tool, select the corresponding model package name from the drop-down list.
Release Original Model Package After Switching	Parameter description: This parameter determines whether to release the resources occupied by the original model package immediately when the model package is switched. Default setting: Selected. Instruction: Once this option is selected, when the Step switches to another model package, the system will immediately release the original model package resources, even if the model package is still in use by other Steps. When this option is not selected, the system will automatically release the original model package only when it is no longer used by any Steps.
Model Package Type	Parameter description: Once a Model Name is selected, the DI Algo Type Translated String will be filled automatically.
Input Batch Size	Parmeter description: The number of images processed during each inference.
GPU ID	Parameter description: This parameter is used to select the device ID of the GPU that will be used for the inference. Tuning instruction: Once you have selected the model name, you can select the GPU ID in the drop-down list of this parameter.

Parameter

Description

Model Manager Tool

Parameter description: This parameter is used to open the deep learning model package management tool and import the deep learning model package. The model package file is a “.dlkpack” file exported by Mech-DLK.
Tuning instruction: Please refer to Deep Learning Model Package Management Tool for the usage.

Model Name

Parameter description: This parameter is used to select the model package that has been imported for this Step.
Tuning instruction: After importing a deep learning model package with the Deep Learning Model Package Management Tool, select the corresponding model package name from the drop-down list.

Release Original Model Package After Switching

Parameter description: This parameter determines whether to release the resources occupied by the original model package immediately when the model package is switched.
Default setting: Selected.
Instruction: Once this option is selected, when the Step switches to another model package, the system will immediately release the original model package resources, even if the model package is still in use by other Steps. When this option is not selected, the system will automatically release the original model package only when it is no longer used by any Steps.

Model Package Type

Parameter description: Once a Model Name is selected, the DI Algo Type Translated String will be filled automatically.

Input Batch Size

Parmeter description: The number of images processed during each inference.

GPU ID

Parameter description: This parameter is used to select the device ID of the GPU that will be used for the inference.
Tuning instruction: Once you have selected the model name, you can select the GPU ID in the drop-down list of this parameter.

Pre-Process

Parameter

Description

ROI File

Parameter description: This parameter is used to set or modify the ROI of the input image.

Tuning instruction: Once the deep learning model is imported, a default ROI will be applied. If you need to edit the ROI, click Open the editor. Edit the ROI in the pop-up Set ROI window, and fill in the ROI name.

Before the inference, please check whether the ROI set here is consistent with the one set in Mech-DLK. If not, the recognition result may be affected.

During the inference, the ROI set during model training, i.e. the default ROI, is usually used. If the position of the object changes in the camera’s field of view, please adjust the ROI.

If you would like to use the default ROI again, please delete the ROI file name below the Open the editor button.

Post-Process

Parameter	Description
Morphological Transformation	Parameter description: When enabled, morphological processing is applied to target object and bin segmentation results. Default value: Disabled.
Morphological Transformation Type	Parameter description: This parameter is used to select the morphological post-processing method for masks. Value list: Dilation, Erosion Dilation: Used to enlarge deep learning masks. When the size of the deep learning mask is smaller than that of the actual target object or bin, the point cloud extracted by the mask may have defects, especially in the edge point cloud. In this case, it is recommended to enable Dilation to increase the mask area to avoid missing extracted point clouds. Erosion: Used to shrink deep learning masks. When the mask coverage of the deep learning model is larger than the actual target object or bin area or contains background noise, it is recommended to enable Erode to shrink the mask coverage to avoid point clouds mixing into non-target areas.
Kernel Size	Parameter description: This parameter is used to set the kernel size of morphological transformation. A larger kernel gives a stronger effect. Default value: 3 px Tuning recommendation: Please adjust the kernel size according to the actual requirement.

Parameter

Description

Morphological Transformation

Parameter description: When enabled, morphological processing is applied to target object and bin segmentation results.
Default value: Disabled.

Morphological Transformation Type

Parameter description: This parameter is used to select the morphological post-processing method for masks.

Value list: Dilation, Erosion

Dilation: Used to enlarge deep learning masks. When the size of the deep learning mask is smaller than that of the actual target object or bin, the point cloud extracted by the mask may have defects, especially in the edge point cloud. In this case, it is recommended to enable Dilation to increase the mask area to avoid missing extracted point clouds.
Erosion: Used to shrink deep learning masks. When the mask coverage of the deep learning model is larger than the actual target object or bin area or contains background noise, it is recommended to enable Erode to shrink the mask coverage to avoid point clouds mixing into non-target areas.

Kernel Size

Parameter description: This parameter is used to set the kernel size of morphological transformation. A larger kernel gives a stronger effect.
Default value: 3 px
Tuning recommendation: Please adjust the kernel size according to the actual requirement.

Visualization Settings

Parameter	Description
Draw Segmentation Mask on Image	Parameter description: This parameter is used to display the segmentation mask on the image. Tuning instructions: Select this option to enable visualization. The segmentation masks are displayed directly on the image, as shown below:

Parameter

Description

Draw Segmentation Mask on Image

Parameter description: This parameter is used to display the segmentation mask on the image.

Tuning instructions: Select this option to enable visualization. The segmentation masks are displayed directly on the image, as shown below:

Is this page helpful?

Leave a feedback

Thanks for your support!

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.

I accept

I don't accept