boofcv.alg.shapes

Class ShapeFittingOps

java.lang.Object

boofcv.alg.shapes.ShapeFittingOps

public class ShapeFittingOps
extends java.lang.Object

Functions for fitting shapes to sequences of points. Points sequences are often found by computing a shape's contour or edge.

See Also:

CannyEdge, BinaryImageOps.contour(GrayU8, boofcv.struct.ConnectRule, GrayS32)

Constructor Summary

Constructors

Constructor and Description
ShapeFittingOps()

Method Summary

Modifier and Type	Method and Description
static FitData<Circle2D_F64>	averageCircle_F64(java.util.List<Point2D_F64> points, GrowQueue_F64 optional, FitData<Circle2D_F64> outputStorage) Computes a circle which has it's center at the mean position of the provided points and radius is equal to the average distance of each point from the center.
static FitData<Circle2D_F64>	averageCircle_I32(java.util.List<Point2D_I32> points, GrowQueue_F64 optional, FitData<Circle2D_F64> outputStorage) Computes a circle which has it's center at the mean position of the provided points and radius is equal to the average distance of each point from the center.
static FitData<EllipseRotated_F64>	fitEllipse_F64(java.util.List<Point2D_F64> points, int iterations, boolean computeError, FitData<EllipseRotated_F64> outputStorage) Computes the best fit ellipse based on minimizing Euclidean distance.
static FitData<EllipseRotated_F64>	fitEllipse_I32(java.util.List<Point2D_I32> points, int iterations, boolean computeError, FitData<EllipseRotated_F64> outputStorage) Convenience function.
static java.util.List<PointIndex_I32>	fitPolygon(java.util.List<Point2D_I32> sequence, boolean loop, double splitFraction, double minimumSideFraction, int iterations) Fits a polygon to the provided sequence of connected points.
static void	indexToPointIndex(java.util.List<Point2D_I32> sequence, GrowQueue_I32 indexes, FastQueue<PointIndex_I32> output) Converts the list of indexes in a sequence into a list of PointIndex_I32.

Methods inherited from class java.lang.Object

equals, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

ShapeFittingOps

public ShapeFittingOps()

Method Detail

fitPolygon

public static java.util.List<PointIndex_I32> fitPolygon(java.util.List<Point2D_I32> sequence,
                                                        boolean loop,
                                                        double splitFraction,
                                                        double minimumSideFraction,
                                                        int iterations)

Fits a polygon to the provided sequence of connected points. The found polygon is returned as a list of vertices. Each point in the original sequence is guaranteed to be within "toleranceDist' of a line segment.

Internally a split-and-merge algorithm is used. See referenced classes for more information. Consider using internal algorithms directly if this function is a performance bottleneck.

Parameters:

sequence - Ordered and connected list of points.

loop - If true the sequence is a connected at both ends, otherwise it is assumed to not be.

splitFraction - A line will be split if a point is more than this fraction of its length away from the line. Try 0.05

minimumSideFraction - The minimum allowed side length as a function of contour length.

iterations - Maximum number of iterations done to improve the fit. Can be 0. Try 50.

Returns:

Vertexes in the fit polygon.

See Also:

SplitMergeLineFitLoop, SplitMergeLineFitSegment

fitEllipse_F64

public static FitData<EllipseRotated_F64> fitEllipse_F64(java.util.List<Point2D_F64> points,
                                                         int iterations,
                                                         boolean computeError,
                                                         FitData<EllipseRotated_F64> outputStorage)

Computes the best fit ellipse based on minimizing Euclidean distance. An estimate is initially provided using algebraic algorithm which is then refined using non-linear optimization. The amount of non-linear optimization can be controlled using 'iterations' parameter. Will work with partial and complete contours of objects.

NOTE: To improve speed, make calls directly to classes in Georegression. Look at the code for details.

Parameters:

points - (Input) Set of unordered points. Not modified.

iterations - Number of iterations used to refine the fit. If set to zero then an algebraic solution is returned.

computeError - If true it will compute the average Euclidean distance error

outputStorage - (Output/Optional) Storage for the ellipse. Can be null.

Returns:

Found ellipse.

fitEllipse_I32

public static FitData<EllipseRotated_F64> fitEllipse_I32(java.util.List<Point2D_I32> points,
                                                         int iterations,
                                                         boolean computeError,
                                                         FitData<EllipseRotated_F64> outputStorage)

Convenience function. Same as fitEllipse_F64(java.util.List, int, boolean,FitData), but converts the set of integer points into floating point points.

Parameters:

points - (Input) Set of unordered points. Not modified.

iterations - Number of iterations used to refine the fit. If set to zero then an algebraic solution is returned.

computeError - If true it will compute the average Euclidean distance error

outputStorage - (Output/Optional) Storage for the ellipse. Can be null

Returns:

Found ellipse.

averageCircle_I32

public static FitData<Circle2D_F64> averageCircle_I32(java.util.List<Point2D_I32> points,
                                                      GrowQueue_F64 optional,
                                                      FitData<Circle2D_F64> outputStorage)

Computes a circle which has it's center at the mean position of the provided points and radius is equal to the average distance of each point from the center. While fast to compute the provided circle is not a best fit circle by any reasonable metric, except for special cases.

Parameters:

points - (Input) Set of unordered points. Not modified.

optional - (Optional) Used internally to store the distance of each point from the center. Can be null.

outputStorage - (Output/Optional) Storage for results. If null then a new circle instance will be returned.

Returns:

The found circle fit.

averageCircle_F64

public static FitData<Circle2D_F64> averageCircle_F64(java.util.List<Point2D_F64> points,
                                                      GrowQueue_F64 optional,
                                                      FitData<Circle2D_F64> outputStorage)

Computes a circle which has it's center at the mean position of the provided points and radius is equal to the average distance of each point from the center. While fast to compute the provided circle is not a best fit circle by any reasonable metric, except for special cases.

Parameters:

points - (Input) Set of unordered points. Not modified.

optional - (Optional) Used internally to store the distance of each point from the center. Can be null.

outputStorage - (Output/Optional) Storage for results. If null then a new circle instance will be returned.

Returns:

The found circle fit.

indexToPointIndex

public static void indexToPointIndex(java.util.List<Point2D_I32> sequence,
                                     GrowQueue_I32 indexes,
                                     FastQueue<PointIndex_I32> output)

Converts the list of indexes in a sequence into a list of PointIndex_I32.

Parameters:

sequence - Sequence of points.

indexes - List of indexes in the sequence.

output - Output list of PointIndex_I32.