Confusion Property
Applies To
Vision Objects: Correlation, Geometric, Polar
Description
Indicates the amount of confusion expected in the image to be searched. This is the highest shape score a feature can get that is not an instance of the feature for which you are searching. (i.e. Will there be patterns in the image which will "confuse" the searching algorithms? To what level?)
Usage
VGet Sequence.Object.Confusion, var
VSet Sequence.Object.Confusion, value
- Sequence
- Name of a sequence or string variable containing a sequence name.
- Object
- Name of an object or string variable containing an object name. The object must exist in the specified sequence.
- var
- Integer variable that will contain the value of the property.
- value
- Integer value or expression for the new value of the property.
Values
Integer number from 1-999 with the higher numbers representing a higher confusion in an image.
Default: 800 - Correlation, Polar, Geometric
Remarks
Both the Confusion property and the Accept property affect the pattern finding speed.
The search continues until the number of the score patterns specified in NumberToFind which are larger than the Confusion value are found. If the score pattern larger than Accept value and smaller than Confusion is found, the search continues for the higher score pattern, and then confirms the pattern search result if there is no higher score pattern. The Confusion property allows the system to abort the search before exploring all possible regions of the image, thus speeding up the process.
Set the Confusion property based on the highest value you expect the "wrong part" to get (plus a margin for error). It is recommended to set The Confusion property greater than or equal to the Accept property. Setting the Confusion property to a high value may increase the time of the search, while setting the property to a low value may decrease the time of the search. However, note that the low Confusion property value may increase the possibility of false detection.
The Confusion property becomes very important when there are multiple features within an image which are very similar. In these cases the proper setting of the Confusion property (i.e. at a high enough level) helps eliminate the wrong features. However, when an image has few features which look the same then the Confusion property can be set a little lower. This can help reduce processing time.
See Also
Accept Property, Correlation Object, Geometric Object, Polar Object, Score Result