Class RegionGrowing3D

Namespace

OpenCvSharp

Assembly

OpenCvSharp.dll

Region Growing algorithm in 3D point cloud.

public class RegionGrowing3D : Algorithm, IDisposable

Inheritance

object

CvObject

CvPtrObject

Algorithm

RegionGrowing3D

Implements

IDisposable

Inherited Members

Algorithm.Write(FileStorage)

Algorithm.Read(FileNode)

Algorithm.Empty

Algorithm.Save(string)

Algorithm.GetDefaultName()

CvPtrObject.RawPtr

CvObject.IsDisposed

CvObject.ptr

CvObject.SetSafeHandle(OpenCvSafeHandle)

CvObject.Dispose()

CvObject.Dispose(bool)

CvObject.DisposeManaged()

CvObject.DisposeUnmanaged()

CvObject.ThrowIfDisposed()

CvObject.CvPtr

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Remarks

The key idea of region growing is to merge the nearest neighbor points that satisfy a certain angle threshold into the same region according to the normal between the two points, so as to achieve the purpose of segmentation.

Properties

CurvatureThreshold

Threshold value of curvature. Default is 0.05. Only points with curvature less than the threshold will be considered to belong to the same region. If the curvature of each point is not set, this option will not work.

public virtual double CurvatureThreshold { get; set; }

Property Value

double

MaxNumberOfNeighbors

The maximum number of neighbors to use including itself. Setting to a non-positive number or default will use the information from nnIdx.

public virtual int MaxNumberOfNeighbors { get; set; }

Property Value

int

MaxSize

The maximum size of region. Setting to a non-positive number or default will be unlimited.

public virtual int MaxSize { get; set; }

Property Value

int

MinSize

The minimum size of region. Setting to a non-positive number or default will be unlimited.

public virtual int MinSize { get; set; }

Property Value

int

NeedSort

Whether the results need to be sorted in descending order by the number of points.

public virtual bool NeedSort { get; set; }

Property Value

bool

NumberOfRegions

The maximum number of regions you want. Setting to a non-positive number or default will be unlimited.

public virtual int NumberOfRegions { get; set; }

Property Value

int

SmoothModeFlag

Whether to use the smoothness mode. Default is true. If true it will check the angle between the normal of the current point and the normal of its neighbor. Otherwise, it will check the angle between the normal of the seed point and the normal of the current neighbor.

public virtual bool SmoothModeFlag { get; set; }

Property Value

bool

SmoothnessThreshold

Threshold value of the angle between normals, in radian. Default is 30(degree)*PI/180.

public virtual double SmoothnessThreshold { get; set; }

Property Value

double

Methods

Create()

Creates a RegionGrowing3D object.

public static RegionGrowing3D Create()

Returns

RegionGrowing3D

RegionGrowing3D instance

GetCurvatures(OutputArray)

Gets the curvature of each point, if set.

public virtual void GetCurvatures(OutputArray curvatures)

Parameters

curvatures OutputArray

GetSeeds(OutputArray)

Gets the seed points.

public virtual void GetSeeds(OutputArray seeds)

Parameters

seeds OutputArray

Segment(out int[][], OutputArray, InputArray, InputArray, InputArray)

Executes segmentation using the Region Growing algorithm.

public virtual int Segment(out int[][] regionsIdx, OutputArray labels, InputArray inputPts, InputArray normals, InputArray nnIdx)

Parameters

regionsIdx int[][]

Index information of all points in each region.

labels OutputArray

The label corresponds to the model number, 0 means it does not belong to any model, range [0, Number of final resultant models obtained].

inputPts InputArray

Original point cloud, support vector<Point3f> and Mat of size Nx3/3xN.

normals InputArray

Normal of each point, support vector<Point3f> and Mat of size Nx3.

nnIdx InputArray

Index information of nearest neighbors of all points. The first nearest neighbor of each point is itself. Support Mat of size NxK. If the information in a row is [0, 2, 1, -5, -1, 4, 7 ... negative number] it will use only non-negative indexes until it meets a negative number or bound of this row i.e. [0, 2, 1].

Returns

int

Number of final resultant regions obtained by segmentation.

SetCurvatures(InputArray)

Sets the curvature of each point. If not available, pass an empty array.

public virtual void SetCurvatures(InputArray curvatures)

Parameters

curvatures InputArray

SetSeeds(InputArray)

Sets the seed points, it will grow according to the seeds. If not set (or set to an empty array), the default method will be used:

1. If the curvature of each point is set, the seeds will be sorted in ascending order of curvatures.
2. Otherwise, the natural order of the point cloud will be used.

public virtual void SetSeeds(InputArray seeds)

Parameters

seeds InputArray