Validate Box Object Masks

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 determines whether the masks of the boxes meet the requirements according to specified rules and outputs a list of the masks that meet the requirements, thus removing objects that do not meet the box features.

Usage Scenario

This Step is usually used for box depalletizing. It is recommended to use this Step after the Mask Segmentation Step to verify the validity of masks.

Input and Output

Input

Input port Data type Description

Mask Image

Image/Color/Mask[]

Mask lists input to this port will be judged as satisfying the requirements. Mask lists are usually obtained from the Deep Learning Model Package Inference Step.

Depth Map

Image/Depth

Depth map of the box

Mask Images

Image/Color/Mask

Masks of the highest layer regions of the boxes.

Output

Output port Data type Description

Eligible Masks

Image/Color/Mask[]

Mask list of qualified boxes.

Parameter Description

Size Consistency

Parameter Description

All Boxes Must Be Same Dimensions

Description: This parameter is used to filter out abnormally sized masks. Once this option is enabled, the dimensions (length, width, height) of all box masks judged to be valid must be consistent within a certain error range.
Default setting: Selected

Box Dimension Error Tolerance

Description: This parameter indicates the maximum allowable deviation (in mm) between the dimensions of the boxes. Only masks with dimensional deviations within this range will be judged valid.
Default value: 30.000 mm

Missing Detection

Parameter Description

All Boxes Must Be Detected

Description: This parameter is used to determine whether there are missed boxes in the current detection result. Once this option is enabled, all existing boxes must be detected. If there are unrecognized boxes, the output of this Step will be empty.
Default setting: Selected

Size Ratio of Remaining Box to Standard Box

Description: This parameter is used to determine the completeness of the box masks. The system compares the actual area of the current box mask with the area of a standard box. If the ratio is below this threshold, the box is considered incomplete and the box will be removed.
Default value: 50%

Show Margin Mask

Description: Once this option is enabled, the gaps in the remaining masks will be displayed.
Default setting: Selected

Margin Width

Description: This parameter is used to set the gap (in mm).
Default value: 11

Rectangularity Validation

Parameter Description

Rectangularity Validation

Description: This parameter is used to verify the rectangularity of the box mask. Rectangularity = Mask area / Min bounding rectangle area of the mask.
Default setting: Selected

Min Rectangularity

Description: This parameter is used to set the rectangularity threshold. Boxes with rectangularity values smaller than this threshold will have their masks removed.
Default value: 90.00%

Filter Type

Description: This parameter is used to specify the handling strategy when a non-rectangle box is detected. Please select according to the actual requirement.

  • RemoveAlllfExistsNonRectangularBox: If there are non-rectangle boxes, this Step has no output.

  • JustRemoveNonRectangularBoxes: If there are non-rectangle boxes, this Step only outputs information of rectangular boxes.

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.