Cog3DAlignPattern.GetTrainedBoundingBox Method

Cog3DAlignPattern GetTrainedBoundingBox Method

Cognex VisionPro

Gets the bounding box of the trained points for this pattern.

Namespace: Cognex.VisionPro3D
Assembly: Cognex.VisionPro3D (in Cognex.VisionPro3D.dll) Version: 65.1.0.0

Syntax

public Cog3DBox GetTrainedBoundingBox()

public Cog3DBox GetTrainedBoundingBox()

Public Function GetTrainedBoundingBox As Cog3DBox

Public Function GetTrainedBoundingBox As Cog3DBox

public:
Cog3DBox^ GetTrainedBoundingBox()

public:
Cog3DBox^ GetTrainedBoundingBox()

Return Value

Type: Cog3DBox
A Cog3DBox that best describes the the bounds of the 3D data points that comprise this trained pattern.

Exceptions

Exception	Condition
CogOperatorNotTrainedException	This pattern operator is not trained.

Remarks

The bounding box is a 3D box that encompasses all of the 3D data points trained within the pattern. The bounding box is similar to the TrainRegion of the pattern in the following ways:

The bounding box contains all of the 3D points that are contained within the train region.
The Z-axis of the bounding box is aligned with the Z-axis of the train region.
The bottom face of the bounding box is coincident with the bottom face of the train region.

The key feature of the bounding box is that five of the six faces have been moved inward -- toward the center of the box -- until they contact the training data. The resulting box is a smaller version of the train region: one that is just large enough to contain the data points, except for the bottom face which remains unmoved.

The bottom face of the box is not moved upward because the 3D training data is usually gathered by a camera that cannot see the bottom surface of the physical scene. The thickness of the observed object cannot be estimated solely from its top surface.