Object that aggregates the learning parameters and produces a ShapeFinder2 classifier
More...
|
| Classifier | Learn (ImagePlane plane, Point2D position, Rect teachWindow, IEnumerable< Point2D > dontCarePoints) |
| | Create a ShapeFinder2 classifier from the input image plane . More...
|
| |
| Classifier | Learn (ImagePlane plane, Point2D position, Rect teachWindow) |
| | Create a ShapeFinder2 classifier from the input image plane . If the input plane has overlay information, then this overlay information will be extracted and used as don't care points. More...
|
| |
| Classifier | Learn (ImagePlane plane, Point2D position, Angle angleOffset, double scaleFactor, Rect teachWindow) |
| | Create a ShapeFinder2 classifier from the input image plane . If the input plane has overlay information, then this overlay information will be extracted and used as don't care points. More...
|
| |
| Classifier | Learn (ImagePlane plane, Point2D position, Angle angleOffset, double scaleFactor, Rect teachWindow, IEnumerable< Point2D > dontCarePoints) |
| | Create a ShapeFinder2 classifier from the input More...
|
| |
|
|
static readonly ValueRange< Angle > | RotationRangeMax = new ValueRange<Angle>(Angle.FromDegrees(-180.0), Angle.FromDegrees(180.0)) |
| | The maximum range of rotations (in degrees) that may be set on the learner.
|
| |
|
static readonly ValueRange< double > | ScaleRangeMax = new ValueRange<double>(0.66, 1.50) |
| | The maximum range of scales that may be set on the learner.
|
| |
|
const int | FeatureCountMin = 10 |
| | Minimum number of features a classifier must have.
|
| |
|
const int | ContrastThresholdMin = 1 |
| | Minimum contrast for features eligible for entering the classifier.
|
| |
|
| ValueRange< Angle > | RotationRange [get, set] |
| | Range of rotations that the classifier should be able to cover. Bigger ranges will cause higher processing time. More...
|
| |
| ValueRange< double > | ScaleRange [get, set] |
| | Range of rotations that the classifier should be able to cover. Bigger ranges will cause higher processing time. Angles are measured in degrees. More...
|
| |
| int | FeatureCount [get, set] |
| | Minimum number of features the result classifier should have More...
|
| |
| int | ContrastThreshold [get, set] |
| | Minimum contrast a feature must have to enter into the classifier. More...
|
| |
| int | MaxCoarseLayerScale [get, set] |
| | Maximum exponent of the scale factor between the coarse layer (used for feature search) and the image. The scale factor is determined by 2^scale, e.g. for MaxCoarseLayerScale = 3, the scale factor is 2^3 = More...
|
| |
| int | ProfileSize [get, set] |
| | Profile size gives the number of profile points to be used for correlation around each feature. With the default value of 1, only the feature locations will be used for the optional correlation steps during a ShapeFinder2 search. When set to values > 1 a number of additional points on a line perpendicular to the edge through the feature will be used. Only odd values are valid here. The spacing of the additional points is controlled by the ProfileDelta property. More...
|
| |
| int | ProfileDelta [get, set] |
| | Distance (in pixels) between adjacent profile points. This value only has an effect if the ProfileSize property is > 1. More...
|
| |
Object that aggregates the learning parameters and produces a ShapeFinder2 classifier
◆ Learn() [1/4]
Create a ShapeFinder2 classifier from the input image plane . If the input plane has overlay information, then this overlay information will be extracted and used as don't care points.
- Parameters
-
| plane | image plane to create the classifier from |
| position | position in the input image at which to extract the classifier |
| teachWindow | teach window for the classifier learning process relative to position |
| angleOffset | initial rotation of the object; this angle will be added to the measured angle of the found object(s) at the end of the search function |
| scaleFactor | initial scale factor of the object; this scale factor will be multiplied with the measured scale factor of the found object(s) at the end of the search function |
- Returns
- the newly created classifier
- Exceptions
-
| ArgumentNullException | If the Parent of the plane parameter is null |
| ObjectDisposedException | If the Parent of plane has already been disposed |
| CvbException | when trying to learn a classifier from an image plane that does not have the data type 8 bits per pixel unsigned, or if an unspecific error occurs |
◆ Learn() [2/4]
Create a ShapeFinder2 classifier from the input
- Parameters
-
| plane | plane in the input image from which to create the classifier |
| position | position in the input image at which to extract the classifier |
| teachWindow | teach window for the classifier learning process relative to position |
| angleOffset | initial rotation of the object; this angle will be added to the measured angle of the found object(s) at the end of the search function |
| scaleFactor | initial scale factor of the object; this scale factor will be multiplied with the measured scale factor of the found object(s) at the end of the search function |
| dontCarePoints | list of points relative to position from which no features for the classifier should be extracted |
; an empty list or null handle is acceptable here if no don't care points are to be used
- Returns
- the newly created classifier
- Exceptions
-
| ArgumentNullException | If the Parent of the plane parameter is null |
| ObjectDisposedException | If the Parent of plane has already been disposed |
| CvbException | when trying to learn a classifier from an image plane that does not have the data type 8 bits per pixel unsigned, or if an unspecific error occurs |
◆ Learn() [3/4]
Create a ShapeFinder2 classifier from the input image plane . If the input plane has overlay information, then this overlay information will be extracted and used as don't care points.
- Parameters
-
| plane | image plane to create the classifier from |
| position | position in the input image at which to extract the classifier |
| teachWindow | teach window for the classifier learning process relative to position |
- Returns
- the newly created classifier
- Exceptions
-
| ArgumentNullException | If the Parent of the plane parameter is null |
| ObjectDisposedException | If the Parent of plane has already been disposed |
| CvbException | when trying to learn a classifier from an image plane that does not have the data type 8 bits per pixel unsigned, or if an unspecific error occurs |
◆ Learn() [4/4]
Create a ShapeFinder2 classifier from the input image plane .
- Parameters
-
| plane | image plane to create the classifier from |
| position | position in the input image at which to extract the classifier |
| teachWindow | teach window for the classifier learning process relative to position |
| dontCarePoints | list of points relative to position from which no features for the classifier should be extracted |
; an empty list or null handle is acceptable here if no don't care points are to be used
- Returns
- the newly created classifier
- Exceptions
-
| ArgumentNullException | If the Parent of the plane parameter is null |
| ObjectDisposedException | If the Parent of plane has already been disposed |
| CvbException | when trying to learn a classifier from an image plane that does not have the data type 8 bits per pixel unsigned, or if an unspecific error occurs |
◆ ContrastThreshold
Minimum contrast a feature must have to enter into the classifier.
- Exceptions
-
| ArgumentOutOfRangeException | If the contrast threshold is below the minimum defined by ContrastThresholdMin |
◆ FeatureCount
Minimum number of features the result classifier should have
- Exceptions
-
| ArgumentOutOfRangeException | If the feature count is below the minimum defined by FeatureCountMin |
◆ MaxCoarseLayerScale
Maximum exponent of the scale factor between the coarse layer (used for feature search) and the image. The scale factor is determined by 2^scale, e.g. for MaxCoarseLayerScale = 3, the scale factor is 2^3 =
- When set to -1 (default) the coarse layer scale will be determined automatically by ShapeFinder2 depending on the image size and number features that were extracted.
- Exceptions
-
| ArgumentOutOfRangeException | If the scale is less than -1 |
◆ ProfileDelta
Distance (in pixels) between adjacent profile points. This value only has an effect if the ProfileSize property is > 1.
- Exceptions
-
| ArgumentOutOfRangeException | If the ProfileDelta is less than 1 |
◆ ProfileSize
Profile size gives the number of profile points to be used for correlation around each feature. With the default value of 1, only the feature locations will be used for the optional correlation steps during a ShapeFinder2 search. When set to values > 1 a number of additional points on a line perpendicular to the edge through the feature will be used. Only odd values are valid here. The spacing of the additional points is controlled by the ProfileDelta property.
- Exceptions
-
| ArgumentOutOfRangeException | If the ProfileSize is less than 1 |
| ArgumentException | If the ProfileSize is not an odd number |
◆ RotationRange
Range of rotations that the classifier should be able to cover. Bigger ranges will cause higher processing time.
- Exceptions
-
| ArgumentOutOfRangeException | If the rotation range to be set exceeds the one defined in RotationRangeMax |
◆ ScaleRange
Range of rotations that the classifier should be able to cover. Bigger ranges will cause higher processing time. Angles are measured in degrees.
- Exceptions
-
| ArgumentOutOfRangeException | If the rotation range to be set exceeds the one defined in RotationRangeMax |
