public Cog3DAlignTrainAlgorithmConstants TrainedAlgorithm { get; }

public Cog3DAlignTrainAlgorithmConstants TrainedAlgorithm { get; }

Public ReadOnly Property TrainedAlgorithm As Cog3DAlignTrainAlgorithmConstants 
	Get

Public ReadOnly Property TrainedAlgorithm As Cog3DAlignTrainAlgorithmConstants 
	Get

public:
property Cog3DAlignTrainAlgorithmConstants TrainedAlgorithm {
	Cog3DAlignTrainAlgorithmConstants get ();
}

public:
property Cog3DAlignTrainAlgorithmConstants TrainedAlgorithm {
	Cog3DAlignTrainAlgorithmConstants get ();
}

Property Value

At runtime 3DAlign will use one of two different algorithms for locating pattern instances. Typcially, the SimplePart algorithm works best on objects that are composed of simple planar surfaces. The ComplexPart algorithm works best on objects that are composed of complex curved surfaces.

The algorithm to use at runtime is chosen using the TrainAlgorithm property. This property must be set before you train your pattern. Since it can be challenging to determine which algorithm will be best for a given set of 3D training data you should usually use the default choice, AutoSelect.

AutoSelect causes 3DAlign to automatically determine which of the two algorithms is likely to work best on the given 3D training data. The automatically-chosen algorithm is then trained and used at runtime. You should verify that the chosen algorithm provides the speed, accuracy, and robustness that you need for your application. If it does not, you can explicitly try the other algorithm.

In a trained pattern you can use the TrainedAlgorithm property to determine which algorithm was trained.