Locator l Positioning Level has only a slight impact on the execution speed. l In applications where accuracy is not critical, decreasing the value can provide a slight improvement in speed. Recognition Level Recognition Level allows you to slightly modify the level of recognition effort. The default set- ting of 5 is the optimized (and recommended) setting for typical applications. Recognition Level can be from 0 to 10. A value of 0 will lead to a faster search that may miss instances that are partly blocked. Con- versely, a value of 10 is useful for finding partly blocked objects in cluttered or noisy images, or for models made up of small features at the Outline Level.
l When changing the recognition level, test your application to find the optimum speed at which the process will still find all necessary objects within the image. l If recognition level is too low (quick) some instances may be ignored. l If recognition level is too high (exhaustive), your application may run too slowly. l Recognition level does not affect positioning accuracy. Model Show Model Name Determines if the model name is displayed in the Vision Window. This is only evaluated if Show Results Graphics is true.
Model Disambiguation Enabled When Model Disambiguation Enabled is set to true (default), the Locator applies dis- ambiguation to discriminate between similar models and between similar hypotheses of a single object. When set to false, the Locator does not apply disambiguation. Model Optimizer Enabled Model Optimizer Enabled specifies if the models can be optimized interactively using the Model Optimizer. When set to true, the models can be optimized. When set to false, the mod- els cannot be modified. To build an optimized Model, you must provide an initial model of the object, and then execute the Locator tool.
While optimization is enabled, each new instance of the object found by the Locator is analyzed and compiled into the optimized model that is currently in progress. Strong features that recur frequently in the analyzed instances are retained in the optimized model. Once the model is considered satisfactory, the optimized model can be saved. ACEUser's Guide, 3.6.x, 18316-000 Rev A Page 571