Class HOGDescriptor

HOG (Histogram-of-Oriented-Gradients) Descriptor and Object Detector

Inheritance

System.Object

DisposableObject

DisposableCvObject

HOGDescriptor

Implements

ICvPtrHolder

Inherited Members

DisposableCvObject.ptr

DisposableCvObject.CvPtr

DisposableObject.DataHandle

DisposableObject.IsDisposed

DisposableObject.IsEnabledDispose

DisposableObject.AllocatedMemory

DisposableObject.AllocatedMemorySize

DisposableObject.Dispose()

DisposableObject.Dispose(Boolean)

DisposableObject.DisposeManaged()

DisposableObject.AllocGCHandle(Object)

DisposableObject.AllocMemory(Int32)

DisposableObject.NotifyMemoryPressure(Int64)

DisposableObject.ThrowIfDisposed()

Namespace: OpenCvSharp

Assembly: OpenCvSharp.dll

Syntax

public class HOGDescriptor : DisposableCvObject, ICvPtrHolder

Constructors

| Improve this Doc View Source

HOGDescriptor()

Default constructor

Declaration

public HOGDescriptor()

| Improve this Doc View Source

HOGDescriptor(Nullable<Size>, Nullable<Size>, Nullable<Size>, Nullable<Size>, Int32, Int32, Double, HistogramNormType, Double, Boolean, Int32)

Creates the HOG descriptor and detector.

Declaration

public HOGDescriptor(Size? winSize = null, Size? blockSize = null, Size? blockStride = null, Size? cellSize = null, int nbins = 9, int derivAperture = 1, double winSigma = -1, HistogramNormType histogramNormType = HistogramNormType.L2Hys, double l2HysThreshold = 0.2, bool gammaCorrection = true, int nlevels = 64)

Parameters

Type	Name	Description
System.Nullable<OpenCvSharp.Size>	winSize	Detection window size. Align to block size and block stride.
System.Nullable<OpenCvSharp.Size>	blockSize	Block size in pixels. Align to cell size. Only (16,16) is supported for now.
System.Nullable<OpenCvSharp.Size>	blockStride	Block stride. It must be a multiple of cell size.
System.Nullable<OpenCvSharp.Size>	cellSize	Cell size. Only (8, 8) is supported for now.
System.Int32	nbins	Number of bins. Only 9 bins per cell are supported for now.
System.Int32	derivAperture
System.Double	winSigma	Gaussian smoothing window parameter.
HistogramNormType	histogramNormType
System.Double	l2HysThreshold	L2-Hys normalization method shrinkage.
System.Boolean	gammaCorrection	Flag to specify whether the gamma correction preprocessing is required or not.
System.Int32	nlevels	Maximum number of detection window increases.

| Improve this Doc View Source

HOGDescriptor(String)

Construct from a file containing HOGDescriptor properties and coefficients for the linear SVM classifier.

Declaration

public HOGDescriptor(string fileName)

Parameters

Type	Name	Description
System.String	fileName	The file name containing HOGDescriptor properties and coefficients for the linear SVM classifier.

Fields

| Improve this Doc View Source

DaimlerPeopleDetector

This field returns 1981 SVM coeffs obtained from daimler's base. To use these coeffs the detection window size should be (48,96)

Declaration

public static readonly float[] DaimlerPeopleDetector

Field Value

Type	Description
System.Single[]

| Improve this Doc View Source

DefaultNlevels

Declaration

public const int DefaultNlevels = 64

Field Value

Type	Description
System.Int32

| Improve this Doc View Source

DefaultPeopleDetector

Returns coefficients of the classifier trained for people detection (for default window size).

Declaration

public static readonly float[] DefaultPeopleDetector

Field Value

Type	Description
System.Single[]

| Improve this Doc View Source

L2Hys

Declaration

public const int L2Hys = 0

Field Value

Type	Description
System.Int32

Properties

| Improve this Doc View Source

BlockSize

Block size in pixels. Align to cell size. Default value is Size(16,16).

Declaration

public Size BlockSize { get; set; }

Property Value

Type	Description
OpenCvSharp.Size

| Improve this Doc View Source

BlockStride

Block stride. It must be a multiple of cell size. Default value is Size(8,8).

Declaration

public Size BlockStride { get; set; }

Property Value

Type	Description
OpenCvSharp.Size

| Improve this Doc View Source

CellSize

Cell size. Default value is Size(8,8).

Declaration

public Size CellSize { get; set; }

Property Value

Type	Description
OpenCvSharp.Size

| Improve this Doc View Source

DerivAperture

Declaration

public int DerivAperture { get; set; }

Property Value

Type	Description
System.Int32

| Improve this Doc View Source

GammaCorrection

Flag to specify whether the gamma correction preprocessing is required or not.

Declaration

public bool GammaCorrection { get; set; }

Property Value

Type	Description
System.Boolean

| Improve this Doc View Source

HistogramNormType

Declaration

public HistogramNormType HistogramNormType { get; set; }

Property Value

Type	Description
HistogramNormType

| Improve this Doc View Source

L2HysThreshold

L2-Hys normalization method shrinkage.

Declaration

public double L2HysThreshold { get; set; }

Property Value

Type	Description
System.Double

| Improve this Doc View Source

Nbins

Number of bins used in the calculation of histogram of gradients. Default value is 9.

Declaration

public int Nbins { get; set; }

Property Value

Type	Description
System.Int32

| Improve this Doc View Source

NLevels

Maximum number of detection window increases. Default value is 64

Declaration

public int NLevels { get; set; }

Property Value

Type	Description
System.Int32

| Improve this Doc View Source

SignedGradient

Indicates signed gradient will be used or not

Declaration

public bool SignedGradient { get; set; }

Property Value

Type	Description
System.Boolean

| Improve this Doc View Source

WinSigma

Gaussian smoothing window parameter.

Declaration

public double WinSigma { get; set; }

Property Value

Type	Description
System.Double

| Improve this Doc View Source

WinSize

Detection window size. Align to block size and block stride. Default value is Size(64,128).

Declaration

public Size WinSize { get; set; }

Property Value

Type	Description
OpenCvSharp.Size

Methods

| Improve this Doc View Source

CheckDetectorSize()

Declaration

public bool CheckDetectorSize()

Returns

Type	Description
System.Boolean

| Improve this Doc View Source

Compute(Mat, Nullable<Size>, Nullable<Size>, Point[])

Computes HOG descriptors of given image.

Declaration

public virtual float[] Compute(Mat img, Size? winStride = null, Size? padding = null, Point[] locations = null)

Parameters

Type	Name	Description
Mat	img	Matrix of the type CV_8U containing an image where HOG features will be calculated.
System.Nullable<OpenCvSharp.Size>	winStride	Window stride. It must be a multiple of block stride.
System.Nullable<OpenCvSharp.Size>	padding	Padding
OpenCvSharp.Point[]	locations	Vector of Point

Returns

Type	Description
System.Single[]	Matrix of the type CV_32F

| Improve this Doc View Source

ComputeGradient(Mat, Mat, Mat, Nullable<Size>, Nullable<Size>)

Computes gradients and quantized gradient orientations.

Declaration

public virtual void ComputeGradient(Mat img, Mat grad, Mat angleOfs, Size? paddingTL = null, Size? paddingBR = null)

Parameters

Type	Name	Description
Mat	img	Matrix contains the image to be computed
Mat	grad	Matrix of type CV_32FC2 contains computed gradients
Mat	angleOfs	Matrix of type CV_8UC2 contains quantized gradient orientations
System.Nullable<OpenCvSharp.Size>	paddingTL	Padding from top-left
System.Nullable<OpenCvSharp.Size>	paddingBR	Padding from bottom-right

| Improve this Doc View Source

Detect(Mat, Double, Nullable<Size>, Nullable<Size>, Point[])

Performs object detection without a multi-scale window.

Declaration

public virtual Point[] Detect(Mat img, double hitThreshold = 0, Size? winStride = null, Size? padding = null, Point[] searchLocations = null)

Parameters

Type	Name	Description
Mat	img	Source image. CV_8UC1 and CV_8UC4 types are supported for now.
System.Double	hitThreshold	Threshold for the distance between features and SVM classifying plane. Usually it is 0 and should be specified in the detector coefficients (as the last free coefficient). But if the free coefficient is omitted (which is allowed), you can specify it manually here.
System.Nullable<OpenCvSharp.Size>	winStride	Window stride. It must be a multiple of block stride.
System.Nullable<OpenCvSharp.Size>	padding	Mock parameter to keep the CPU interface compatibility. It must be (0,0).
OpenCvSharp.Point[]	searchLocations

Returns

Type	Description
OpenCvSharp.Point[]	Left-top corner points of detected objects boundaries.

| Improve this Doc View Source

Detect(Mat, out Double[], Double, Nullable<Size>, Nullable<Size>, Point[])

Performs object detection without a multi-scale window.

Declaration

public virtual Point[] Detect(Mat img, out double[] weights, double hitThreshold = 0, Size? winStride = null, Size? padding = null, Point[] searchLocations = null)

Parameters

Type	Name	Description
Mat	img	Source image. CV_8UC1 and CV_8UC4 types are supported for now.
System.Double[]	weights
System.Double	hitThreshold	Threshold for the distance between features and SVM classifying plane. Usually it is 0 and should be specfied in the detector coefficients (as the last free coefficient). But if the free coefficient is omitted (which is allowed), you can specify it manually here.
System.Nullable<OpenCvSharp.Size>	winStride	Window stride. It must be a multiple of block stride.
System.Nullable<OpenCvSharp.Size>	padding	Mock parameter to keep the CPU interface compatibility. It must be (0,0).
OpenCvSharp.Point[]	searchLocations

Returns

Type	Description
OpenCvSharp.Point[]	Left-top corner points of detected objects boundaries.

| Improve this Doc View Source

DetectMultiScale(Mat, Double, Nullable<Size>, Nullable<Size>, Double, Int32)

Performs object detection with a multi-scale window.

Declaration

public virtual Rect[] DetectMultiScale(Mat img, double hitThreshold = 0, Size? winStride = null, Size? padding = null, double scale = 1.05, int groupThreshold = 2)

Parameters

Type	Name	Description
Mat	img	Source image. CV_8UC1 and CV_8UC4 types are supported for now.
System.Double	hitThreshold	Threshold for the distance between features and SVM classifying plane.
System.Nullable<OpenCvSharp.Size>	winStride	Window stride. It must be a multiple of block stride.
System.Nullable<OpenCvSharp.Size>	padding	Mock parameter to keep the CPU interface compatibility. It must be (0,0).
System.Double	scale	Coefficient of the detection window increase.
System.Int32	groupThreshold	Coefficient to regulate the similarity threshold. When detected, some objects can be covered by many rectangles. 0 means not to perform grouping.

Returns

Type	Description
OpenCvSharp.Rect[]	Detected objects boundaries.

| Improve this Doc View Source

DetectMultiScale(Mat, out Double[], Double, Nullable<Size>, Nullable<Size>, Double, Int32)

Performs object detection with a multi-scale window.

Declaration

public virtual Rect[] DetectMultiScale(Mat img, out double[] foundWeights, double hitThreshold = 0, Size? winStride = null, Size? padding = null, double scale = 1.05, int groupThreshold = 2)

Parameters

Type	Name	Description
Mat	img	Source image. CV_8UC1 and CV_8UC4 types are supported for now.
System.Double[]	foundWeights
System.Double	hitThreshold	Threshold for the distance between features and SVM classifying plane.
System.Nullable<OpenCvSharp.Size>	winStride	Window stride. It must be a multiple of block stride.
System.Nullable<OpenCvSharp.Size>	padding	Mock parameter to keep the CPU interface compatibility. It must be (0,0).
System.Double	scale	Coefficient of the detection window increase.
System.Int32	groupThreshold	Coefficient to regulate the similarity threshold. When detected, some objects can be covered by many rectangles. 0 means not to perform grouping.

Returns

Type	Description
OpenCvSharp.Rect[]	Detected objects boundaries.

| Improve this Doc View Source

DetectMultiScaleROI(Mat, out Rect[], out DetectionROI[], Double, Int32)

evaluate specified ROI and return confidence value for each location in multiple scales

Declaration

public void DetectMultiScaleROI(Mat img, out Rect[] foundLocations, out DetectionROI[] locations, double hitThreshold = 0, int groupThreshold = 0)

Parameters

Type	Name	Description
Mat	img	Matrix of the type CV_8U or CV_8UC3 containing an image where objects are detected.
OpenCvSharp.Rect[]	foundLocations	Vector of rectangles where each rectangle contains the detected object.
DetectionROI[]	locations	Vector of DetectionROI
System.Double	hitThreshold	Threshold for the distance between features and SVM classifying plane. Usually it is 0 and should be specified in the detector coefficients (as the last free coefficient). But if the free coefficient is omitted (which is allowed), you can specify it manually here.
System.Int32	groupThreshold	Minimum possible number of rectangles minus 1. The threshold is used in a group of rectangles to retain it.

| Improve this Doc View Source

DetectROI(Mat, Point[], out Point[], out Double[], Double, Nullable<Size>, Nullable<Size>)

evaluate specified ROI and return confidence value for each location

Declaration

public void DetectROI(Mat img, Point[] locations, out Point[] foundLocations, out double[] confidences, double hitThreshold = 0, Size? winStride = null, Size? padding = null)

Parameters

Type	Name	Description
Mat	img	Matrix of the type CV_8U or CV_8UC3 containing an image where objects are detected.
OpenCvSharp.Point[]	locations	Vector of Point
OpenCvSharp.Point[]	foundLocations	Vector of Point where each Point is detected object's top-left point.
System.Double[]	confidences	confidences
System.Double	hitThreshold	Threshold for the distance between features and SVM classifying plane. Usually it is 0 and should be specified in the detector coefficients (as the last free coefficient). But if the free coefficient is omitted (which is allowed), you can specify it manually here
System.Nullable<OpenCvSharp.Size>	winStride	winStride
System.Nullable<OpenCvSharp.Size>	padding	padding

| Improve this Doc View Source

DisposeUnmanaged()

Releases unmanaged resources

Declaration

protected override void DisposeUnmanaged()

Overrides

DisposableCvObject.DisposeUnmanaged()

| Improve this Doc View Source

GetDaimlerPeopleDetector()

This method returns 1981 SVM coeffs obtained from daimler's base. To use these coeffs the detection window size should be (48,96)

Declaration

public static float[] GetDaimlerPeopleDetector()

Returns

Type	Description
System.Single[]

| Improve this Doc View Source

GetDefaultPeopleDetector()

Returns coefficients of the classifier trained for people detection (for default window size).

Declaration

public static float[] GetDefaultPeopleDetector()

Returns

Type	Description
System.Single[]

| Improve this Doc View Source

GetDescriptorSize()

Declaration

public int GetDescriptorSize()

Returns

Type	Description
System.Int32

| Improve this Doc View Source

GetWinSigma()

Declaration

public double GetWinSigma()

Returns

Type	Description
System.Double

| Improve this Doc View Source

GroupRectangles(out Rect[], out Double[], Int32, Double)

Groups the object candidate rectangles.

Declaration

public void GroupRectangles(out Rect[] rectList, out double[] weights, int groupThreshold, double eps)

Parameters

Type	Name	Description
OpenCvSharp.Rect[]	rectList	Input/output vector of rectangles. Output vector includes retained and grouped rectangles. (The Python list is not modified in place.)
System.Double[]	weights	Input/output vector of weights of rectangles. Output vector includes weights of retained and grouped rectangles. (The Python list is not modified in place.)
System.Int32	groupThreshold	Minimum possible number of rectangles minus 1. The threshold is used in a group of rectangles to retain it.
System.Double	eps	Relative difference between sides of the rectangles to merge them into a group.

| Improve this Doc View Source

Load(String, String)

loads HOGDescriptor parameters and coefficients for the linear SVM classifier from a file.

Declaration

public virtual bool Load(string fileName, string objName = null)

Parameters

Type	Name	Description
System.String	fileName	Path of the file to read.
System.String	objName	The optional name of the node to read (if empty, the first top-level node will be used).

Returns

Type	Description
System.Boolean

| Improve this Doc View Source

Save(String, String)

saves HOGDescriptor parameters and coefficients for the linear SVM classifier to a file

Declaration

public virtual void Save(string fileName, string objName = null)

Parameters

Type	Name	Description
System.String	fileName	File name
System.String	objName	Object name

| Improve this Doc View Source

SetSVMDetector(Single[])

Sets coefficients for the linear SVM classifier.

Declaration

public virtual void SetSVMDetector(float[] svmDetector)

Parameters

Type	Name	Description
System.Single[]	svmDetector	coefficients for the linear SVM classifier.

Implements

ICvPtrHolder