OpenCV  3.4.2
Open Source Computer Vision
Namespaces | Classes | Enumerations | Functions
Structural Analysis and Shape Descriptors

Namespaces

 cv::traits
 

Classes

class  cv::Moments
 struct returned by cv::moments More...
 

Enumerations

enum  cv::ConnectedComponentsAlgorithmsTypes {
  cv::CCL_WU = 0,
  cv::CCL_DEFAULT = -1,
  cv::CCL_GRANA = 1
}
 connected components algorithm More...
 
enum  cv::ConnectedComponentsTypes {
  cv::CC_STAT_LEFT = 0,
  cv::CC_STAT_TOP = 1,
  cv::CC_STAT_WIDTH = 2,
  cv::CC_STAT_HEIGHT = 3,
  cv::CC_STAT_AREA = 4,
  cv::CC_STAT_MAX = 5
}
 connected components algorithm output formats More...
 
enum  cv::ContourApproximationModes {
  cv::CHAIN_APPROX_NONE = 1,
  cv::CHAIN_APPROX_SIMPLE = 2,
  cv::CHAIN_APPROX_TC89_L1 = 3,
  cv::CHAIN_APPROX_TC89_KCOS = 4
}
 the contour approximation algorithm More...
 
enum  cv::RectanglesIntersectTypes {
  cv::INTERSECT_NONE = 0,
  cv::INTERSECT_PARTIAL = 1,
  cv::INTERSECT_FULL = 2
}
 
enum  cv::RetrievalModes {
  cv::RETR_EXTERNAL = 0,
  cv::RETR_LIST = 1,
  cv::RETR_CCOMP = 2,
  cv::RETR_TREE = 3,
  cv::RETR_FLOODFILL = 4
}
 mode of the contour retrieval algorithm More...
 
enum  cv::ShapeMatchModes {
  cv::CONTOURS_MATCH_I1 =1,
  cv::CONTOURS_MATCH_I2 =2,
  cv::CONTOURS_MATCH_I3 =3
}
 Shape matching methods. More...
 

Functions

void cv::approxPolyDP (InputArray curve, OutputArray approxCurve, double epsilon, bool closed)
 Approximates a polygonal curve(s) with the specified precision. More...
 
double cv::arcLength (InputArray curve, bool closed)
 Calculates a contour perimeter or a curve length. More...
 
Rect cv::boundingRect (InputArray points)
 Calculates the up-right bounding rectangle of a point set. More...
 
void cv::boxPoints (RotatedRect box, OutputArray points)
 Finds the four vertices of a rotated rect. Useful to draw the rotated rectangle. More...
 
int cv::connectedComponents (InputArray image, OutputArray labels, int connectivity, int ltype, int ccltype)
 computes the connected components labeled image of boolean image More...
 
int cv::connectedComponents (InputArray image, OutputArray labels, int connectivity=8, int ltype=CV_32S)
 
int cv::connectedComponentsWithStats (InputArray image, OutputArray labels, OutputArray stats, OutputArray centroids, int connectivity, int ltype, int ccltype)
 computes the connected components labeled image of boolean image and also produces a statistics output for each label More...
 
int cv::connectedComponentsWithStats (InputArray image, OutputArray labels, OutputArray stats, OutputArray centroids, int connectivity=8, int ltype=CV_32S)
 
double cv::contourArea (InputArray contour, bool oriented=false)
 Calculates a contour area. More...
 
void cv::convexHull (InputArray points, OutputArray hull, bool clockwise=false, bool returnPoints=true)
 Finds the convex hull of a point set. More...
 
void cv::convexityDefects (InputArray contour, InputArray convexhull, OutputArray convexityDefects)
 Finds the convexity defects of a contour. More...
 
void cv::findContours (InputOutputArray image, OutputArrayOfArrays contours, OutputArray hierarchy, int mode, int method, Point offset=Point())
 Finds contours in a binary image. More...
 
void cv::findContours (InputOutputArray image, OutputArrayOfArrays contours, int mode, int method, Point offset=Point())
 
RotatedRect cv::fitEllipse (InputArray points)
 Fits an ellipse around a set of 2D points. More...
 
RotatedRect cv::fitEllipseAMS (InputArray points)
 Fits an ellipse around a set of 2D points. More...
 
RotatedRect cv::fitEllipseDirect (InputArray points)
 Fits an ellipse around a set of 2D points. More...
 
void cv::fitLine (InputArray points, OutputArray line, int distType, double param, double reps, double aeps)
 Fits a line to a 2D or 3D point set. More...
 
void cv::HuMoments (const Moments &moments, double hu[7])
 Calculates seven Hu invariants. More...
 
void cv::HuMoments (const Moments &m, OutputArray hu)
 
float cv::intersectConvexConvex (InputArray _p1, InputArray _p2, OutputArray _p12, bool handleNested=true)
 finds intersection of two convex polygons More...
 
bool cv::isContourConvex (InputArray contour)
 Tests a contour convexity. More...
 
double cv::matchShapes (InputArray contour1, InputArray contour2, int method, double parameter)
 Compares two shapes. More...
 
RotatedRect cv::minAreaRect (InputArray points)
 Finds a rotated rectangle of the minimum area enclosing the input 2D point set. More...
 
void cv::minEnclosingCircle (InputArray points, Point2f &center, float &radius)
 Finds a circle of the minimum area enclosing a 2D point set. More...
 
double cv::minEnclosingTriangle (InputArray points, OutputArray triangle)
 Finds a triangle of minimum area enclosing a 2D point set and returns its area. More...
 
Moments cv::moments (InputArray array, bool binaryImage=false)
 Calculates all of the moments up to the third order of a polygon or rasterized shape. More...
 
double cv::pointPolygonTest (InputArray contour, Point2f pt, bool measureDist)
 Performs a point-in-contour test. More...
 
int cv::rotatedRectangleIntersection (const RotatedRect &rect1, const RotatedRect &rect2, OutputArray intersectingRegion)
 Finds out if there is any intersection between two rotated rectangles. More...
 

Detailed Description

Enumeration Type Documentation

§ ConnectedComponentsAlgorithmsTypes

connected components algorithm

Enumerator
CCL_WU 
Python: cv.CCL_WU

SAUF algorithm for 8-way connectivity, SAUF algorithm for 4-way connectivity.

CCL_DEFAULT 
Python: cv.CCL_DEFAULT

BBDT algorithm for 8-way connectivity, SAUF algorithm for 4-way connectivity.

CCL_GRANA 
Python: cv.CCL_GRANA

BBDT algorithm for 8-way connectivity, SAUF algorithm for 4-way connectivity.

§ ConnectedComponentsTypes

connected components algorithm output formats

Enumerator
CC_STAT_LEFT 
Python: cv.CC_STAT_LEFT

The leftmost (x) coordinate which is the inclusive start of the bounding box in the horizontal direction.

CC_STAT_TOP 
Python: cv.CC_STAT_TOP

The topmost (y) coordinate which is the inclusive start of the bounding box in the vertical direction.

CC_STAT_WIDTH 
Python: cv.CC_STAT_WIDTH

The horizontal size of the bounding box.

CC_STAT_HEIGHT 
Python: cv.CC_STAT_HEIGHT

The vertical size of the bounding box.

CC_STAT_AREA 
Python: cv.CC_STAT_AREA

The total area (in pixels) of the connected component.

CC_STAT_MAX 
Python: cv.CC_STAT_MAX

§ ContourApproximationModes

the contour approximation algorithm

Enumerator
CHAIN_APPROX_NONE 
Python: cv.CHAIN_APPROX_NONE

stores absolutely all the contour points. That is, any 2 subsequent points (x1,y1) and (x2,y2) of the contour will be either horizontal, vertical or diagonal neighbors, that is, max(abs(x1-x2),abs(y2-y1))==1.

CHAIN_APPROX_SIMPLE 
Python: cv.CHAIN_APPROX_SIMPLE

compresses horizontal, vertical, and diagonal segments and leaves only their end points. For example, an up-right rectangular contour is encoded with 4 points.

CHAIN_APPROX_TC89_L1 
Python: cv.CHAIN_APPROX_TC89_L1

applies one of the flavors of the Teh-Chin chain approximation algorithm [181]

CHAIN_APPROX_TC89_KCOS 
Python: cv.CHAIN_APPROX_TC89_KCOS

applies one of the flavors of the Teh-Chin chain approximation algorithm [181]

§ RectanglesIntersectTypes

types of intersection between rectangles

Enumerator
INTERSECT_NONE 
Python: cv.INTERSECT_NONE

No intersection.

INTERSECT_PARTIAL 
Python: cv.INTERSECT_PARTIAL

There is a partial intersection.

INTERSECT_FULL 
Python: cv.INTERSECT_FULL

One of the rectangle is fully enclosed in the other.

§ RetrievalModes

mode of the contour retrieval algorithm

Enumerator
RETR_EXTERNAL 
Python: cv.RETR_EXTERNAL

retrieves only the extreme outer contours. It sets hierarchy[i][2]=hierarchy[i][3]=-1 for all the contours.

RETR_LIST 
Python: cv.RETR_LIST

retrieves all of the contours without establishing any hierarchical relationships.

RETR_CCOMP 
Python: cv.RETR_CCOMP

retrieves all of the contours and organizes them into a two-level hierarchy. At the top level, there are external boundaries of the components. At the second level, there are boundaries of the holes. If there is another contour inside a hole of a connected component, it is still put at the top level.

RETR_TREE 
Python: cv.RETR_TREE

retrieves all of the contours and reconstructs a full hierarchy of nested contours.

RETR_FLOODFILL 
Python: cv.RETR_FLOODFILL

§ ShapeMatchModes

Shape matching methods.

\(A\) denotes object1, \(B\) denotes object2

\(\begin{array}{l} m^A_i = \mathrm{sign} (h^A_i) \cdot \log{h^A_i} \\ m^B_i = \mathrm{sign} (h^B_i) \cdot \log{h^B_i} \end{array}\)

and \(h^A_i, h^B_i\) are the Hu moments of \(A\) and \(B\) , respectively.

Enumerator
CONTOURS_MATCH_I1 
Python: cv.CONTOURS_MATCH_I1

\[I_1(A,B) = \sum _{i=1...7} \left | \frac{1}{m^A_i} - \frac{1}{m^B_i} \right |\]

CONTOURS_MATCH_I2 
Python: cv.CONTOURS_MATCH_I2

\[I_2(A,B) = \sum _{i=1...7} \left | m^A_i - m^B_i \right |\]

CONTOURS_MATCH_I3 
Python: cv.CONTOURS_MATCH_I3

\[I_3(A,B) = \max _{i=1...7} \frac{ \left| m^A_i - m^B_i \right| }{ \left| m^A_i \right| }\]

Function Documentation

§ approxPolyDP()

void cv::approxPolyDP ( InputArray  curve,
OutputArray  approxCurve,
double  epsilon,
bool  closed 
)
Python:
approxCurve=cv.approxPolyDP(curve, epsilon, closed[, approxCurve])

Approximates a polygonal curve(s) with the specified precision.

The function cv::approxPolyDP approximates a curve or a polygon with another curve/polygon with less vertices so that the distance between them is less or equal to the specified precision. It uses the Douglas-Peucker algorithm http://en.wikipedia.org/wiki/Ramer-Douglas-Peucker_algorithm

Parameters
curveInput vector of a 2D point stored in std::vector or Mat
approxCurveResult of the approximation. The type should match the type of the input curve.
epsilonParameter specifying the approximation accuracy. This is the maximum distance between the original curve and its approximation.
closedIf true, the approximated curve is closed (its first and last vertices are connected). Otherwise, it is not closed.
Examples:
contours2.cpp.

§ arcLength()

double cv::arcLength ( InputArray  curve,
bool  closed 
)
Python:
retval=cv.arcLength(curve, closed)

Calculates a contour perimeter or a curve length.

The function computes a curve length or a closed contour perimeter.

Parameters
curveInput vector of 2D points, stored in std::vector or Mat.
closedFlag indicating whether the curve is closed or not.

§ boundingRect()

Rect cv::boundingRect ( InputArray  points)
Python:
retval=cv.boundingRect(points)

Calculates the up-right bounding rectangle of a point set.

The function calculates and returns the minimal up-right bounding rectangle for the specified point set.

Parameters
pointsInput 2D point set, stored in std::vector or Mat.

§ boxPoints()

void cv::boxPoints ( RotatedRect  box,
OutputArray  points 
)
Python:
points=cv.boxPoints(box[, points])

Finds the four vertices of a rotated rect. Useful to draw the rotated rectangle.

The function finds the four vertices of a rotated rectangle. This function is useful to draw the rectangle. In C++, instead of using this function, you can directly use RotatedRect::points method. Please visit the tutorial on Creating Bounding rotated boxes and ellipses for contours for more information.

Parameters
boxThe input rotated rectangle. It may be the output of
pointsThe output array of four vertices of rectangles.

§ connectedComponents() [1/2]

int cv::connectedComponents ( InputArray  image,
OutputArray  labels,
int  connectivity,
int  ltype,
int  ccltype 
)
Python:
retval, labels=cv.connectedComponents(image[, labels[, connectivity[, ltype]]])
retval, labels=cv.connectedComponentsWithAlgorithm(image, connectivity, ltype, ccltype[, labels])

computes the connected components labeled image of boolean image

image with 4 or 8 way connectivity - returns N, the total number of labels [0, N-1] where 0 represents the background label. ltype specifies the output label image type, an important consideration based on the total number of labels or alternatively the total number of pixels in the source image. ccltype specifies the connected components labeling algorithm to use, currently Grana (BBDT) and Wu's (SAUF) algorithms are supported, see the ConnectedComponentsAlgorithmsTypes for details. Note that SAUF algorithm forces a row major ordering of labels while BBDT does not. This function uses parallel version of both Grana and Wu's algorithms if at least one allowed parallel framework is enabled and if the rows of the image are at least twice the number returned by getNumberOfCPUs.

Parameters
imagethe 8-bit single-channel image to be labeled
labelsdestination labeled image
connectivity8 or 4 for 8-way or 4-way connectivity respectively
ltypeoutput image label type. Currently CV_32S and CV_16U are supported.
ccltypeconnected components algorithm type (see the ConnectedComponentsAlgorithmsTypes).

§ connectedComponents() [2/2]

int cv::connectedComponents ( InputArray  image,
OutputArray  labels,
int  connectivity = 8,
int  ltype = CV_32S 
)
Python:
retval, labels=cv.connectedComponents(image[, labels[, connectivity[, ltype]]])
retval, labels=cv.connectedComponentsWithAlgorithm(image, connectivity, ltype, ccltype[, labels])

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters
imagethe 8-bit single-channel image to be labeled
labelsdestination labeled image
connectivity8 or 4 for 8-way or 4-way connectivity respectively
ltypeoutput image label type. Currently CV_32S and CV_16U are supported.

§ connectedComponentsWithStats() [1/2]

int cv::connectedComponentsWithStats ( InputArray  image,
OutputArray  labels,
OutputArray  stats,
OutputArray  centroids,
int  connectivity,
int  ltype,
int  ccltype 
)
Python:
retval, labels, stats, centroids=cv.connectedComponentsWithStats(image[, labels[, stats[, centroids[, connectivity[, ltype]]]]])
retval, labels, stats, centroids=cv.connectedComponentsWithStatsWithAlgorithm(image, connectivity, ltype, ccltype[, labels[, stats[, centroids]]])

computes the connected components labeled image of boolean image and also produces a statistics output for each label

image with 4 or 8 way connectivity - returns N, the total number of labels [0, N-1] where 0 represents the background label. ltype specifies the output label image type, an important consideration based on the total number of labels or alternatively the total number of pixels in the source image. ccltype specifies the connected components labeling algorithm to use, currently Grana's (BBDT) and Wu's (SAUF) algorithms are supported, see the ConnectedComponentsAlgorithmsTypes for details. Note that SAUF algorithm forces a row major ordering of labels while BBDT does not. This function uses parallel version of both Grana and Wu's algorithms (statistics included) if at least one allowed parallel framework is enabled and if the rows of the image are at least twice the number returned by getNumberOfCPUs.

Parameters
imagethe 8-bit single-channel image to be labeled
labelsdestination labeled image
statsstatistics output for each label, including the background label, see below for available statistics. Statistics are accessed via stats(label, COLUMN) where COLUMN is one of ConnectedComponentsTypes. The data type is CV_32S.
centroidscentroid output for each label, including the background label. Centroids are accessed via centroids(label, 0) for x and centroids(label, 1) for y. The data type CV_64F.
connectivity8 or 4 for 8-way or 4-way connectivity respectively
ltypeoutput image label type. Currently CV_32S and CV_16U are supported.
ccltypeconnected components algorithm type (see ConnectedComponentsAlgorithmsTypes).

§ connectedComponentsWithStats() [2/2]

int cv::connectedComponentsWithStats ( InputArray  image,
OutputArray  labels,
OutputArray  stats,
OutputArray  centroids,
int  connectivity = 8,
int  ltype = CV_32S 
)
Python:
retval, labels, stats, centroids=cv.connectedComponentsWithStats(image[, labels[, stats[, centroids[, connectivity[, ltype]]]]])
retval, labels, stats, centroids=cv.connectedComponentsWithStatsWithAlgorithm(image, connectivity, ltype, ccltype[, labels[, stats[, centroids]]])

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Parameters
imagethe 8-bit single-channel image to be labeled
labelsdestination labeled image
statsstatistics output for each label, including the background label, see below for available statistics. Statistics are accessed via stats(label, COLUMN) where COLUMN is one of ConnectedComponentsTypes. The data type is CV_32S.
centroidscentroid output for each label, including the background label. Centroids are accessed via centroids(label, 0) for x and centroids(label, 1) for y. The data type CV_64F.
connectivity8 or 4 for 8-way or 4-way connectivity respectively
ltypeoutput image label type. Currently CV_32S and CV_16U are supported.

§ contourArea()

double cv::contourArea ( InputArray  contour,
bool  oriented = false 
)
Python:
retval=cv.contourArea(contour[, oriented])

Calculates a contour area.

The function computes a contour area. Similarly to moments , the area is computed using the Green formula. Thus, the returned area and the number of non-zero pixels, if you draw the contour using drawContours or fillPoly , can be different. Also, the function will most certainly give a wrong results for contours with self-intersections.

Example:

vector<Point> contour;
contour.push_back(Point2f(0, 0));
contour.push_back(Point2f(10, 0));
contour.push_back(Point2f(10, 10));
contour.push_back(Point2f(5, 4));
double area0 = contourArea(contour);
vector<Point> approx;
approxPolyDP(contour, approx, 5, true);
double area1 = contourArea(approx);
cout << "area0 =" << area0 << endl <<
"area1 =" << area1 << endl <<
"approx poly vertices" << approx.size() << endl;
Parameters
contourInput vector of 2D points (contour vertices), stored in std::vector or Mat.
orientedOriented area flag. If it is true, the function returns a signed area value, depending on the contour orientation (clockwise or counter-clockwise). Using this feature you can determine orientation of a contour by taking the sign of an area. By default, the parameter is false, which means that the absolute value is returned.
Examples:
segment_objects.cpp.

§ convexHull()

void cv::convexHull ( InputArray  points,
OutputArray  hull,
bool  clockwise = false,
bool  returnPoints = true 
)
Python:
hull=cv.convexHull(points[, hull[, clockwise[, returnPoints]]])

Finds the convex hull of a point set.

The function cv::convexHull finds the convex hull of a 2D point set using the Sklansky's algorithm [169] that has O(N logN) complexity in the current implementation.

Parameters
pointsInput 2D point set, stored in std::vector or Mat.
hullOutput convex hull. It is either an integer vector of indices or vector of points. In the first case, the hull elements are 0-based indices of the convex hull points in the original array (since the set of convex hull points is a subset of the original point set). In the second case, hull elements are the convex hull points themselves.
clockwiseOrientation flag. If it is true, the output convex hull is oriented clockwise. Otherwise, it is oriented counter-clockwise. The assumed coordinate system has its X axis pointing to the right, and its Y axis pointing upwards.
returnPointsOperation flag. In case of a matrix, when the flag is true, the function returns convex hull points. Otherwise, it returns indices of the convex hull points. When the output array is std::vector, the flag is ignored, and the output depends on the type of the vector: std::vector<int> implies returnPoints=false, std::vector<Point> implies returnPoints=true.
Note
points and hull should be different arrays, inplace processing isn't supported.
Examples:
convexhull.cpp.

§ convexityDefects()

void cv::convexityDefects ( InputArray  contour,
InputArray  convexhull,
OutputArray  convexityDefects 
)
Python:
convexityDefects=cv.convexityDefects(contour, convexhull[, convexityDefects])

Finds the convexity defects of a contour.

The figure below displays convexity defects of a hand contour:

defects.png
image
Parameters
contourInput contour.
convexhullConvex hull obtained using convexHull that should contain indices of the contour points that make the hull.
convexityDefectsThe output vector of convexity defects. In C++ and the new Python/Java interface each convexity defect is represented as 4-element integer vector (a.k.a. Vec4i): (start_index, end_index, farthest_pt_index, fixpt_depth), where indices are 0-based indices in the original contour of the convexity defect beginning, end and the farthest point, and fixpt_depth is fixed-point approximation (with 8 fractional bits) of the distance between the farthest contour point and the hull. That is, to get the floating-point value of the depth will be fixpt_depth/256.0.

§ findContours() [1/2]

void cv::findContours ( InputOutputArray  image,
OutputArrayOfArrays  contours,
OutputArray  hierarchy,
int  mode,
int  method,
Point  offset = Point() 
)
Python:
image, contours, hierarchy=cv.findContours(image, mode, method[, contours[, hierarchy[, offset]]])

Finds contours in a binary image.

The function retrieves contours from the binary image using the algorithm [173] . The contours are a useful tool for shape analysis and object detection and recognition. See squares.cpp in the OpenCV sample directory.

Note
Since opencv 3.2 source image is not modified by this function.
Parameters
imageSource, an 8-bit single-channel image. Non-zero pixels are treated as 1's. Zero pixels remain 0's, so the image is treated as binary . You can use compare, inRange, threshold , adaptiveThreshold, Canny, and others to create a binary image out of a grayscale or color one. If mode equals to RETR_CCOMP or RETR_FLOODFILL, the input can also be a 32-bit integer image of labels (CV_32SC1).
contoursDetected contours. Each contour is stored as a vector of points (e.g. std::vector<std::vector<cv::Point> >).
hierarchyOptional output vector (e.g. std::vector<cv::Vec4i>), containing information about the image topology. It has as many elements as the number of contours. For each i-th contour contours[i], the elements hierarchy[i][0] , hierarchy[i][1] , hierarchy[i][2] , and hierarchy[i][3] are set to 0-based indices in contours of the next and previous contours at the same hierarchical level, the first child contour and the parent contour, respectively. If for the contour i there are no next, previous, parent, or nested contours, the corresponding elements of hierarchy[i] will be negative.
modeContour retrieval mode, see RetrievalModes
methodContour approximation method, see ContourApproximationModes
offsetOptional offset by which every contour point is shifted. This is useful if the contours are extracted from the image ROI and then they should be analyzed in the whole image context.
Examples:
contours2.cpp, fitellipse.cpp, segment_objects.cpp, shape_example.cpp, and watershed.cpp.

§ findContours() [2/2]

void cv::findContours ( InputOutputArray  image,
OutputArrayOfArrays  contours,
int  mode,
int  method,
Point  offset = Point() 
)
Python:
image, contours, hierarchy=cv.findContours(image, mode, method[, contours[, hierarchy[, offset]]])

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

§ fitEllipse()

RotatedRect cv::fitEllipse ( InputArray  points)
Python:
retval=cv.fitEllipse(points)

Fits an ellipse around a set of 2D points.

The function calculates the ellipse that fits (in a least-squares sense) a set of 2D points best of all. It returns the rotated rectangle in which the ellipse is inscribed. The first algorithm described by [59] is used. Developer should keep in mind that it is possible that the returned ellipse/rotatedRect data contains negative indices, due to the data points being close to the border of the containing Mat element.

Parameters
pointsInput 2D point set, stored in std::vector<> or Mat
Examples:
fitellipse.cpp.

§ fitEllipseAMS()

RotatedRect cv::fitEllipseAMS ( InputArray  points)
Python:
retval=cv.fitEllipseAMS(points)

Fits an ellipse around a set of 2D points.

The function calculates the ellipse that fits a set of 2D points. It returns the rotated rectangle in which the ellipse is inscribed. The Approximate Mean Square (AMS) proposed by [180] is used.

For an ellipse, this basis set is \( \chi= \left(x^2, x y, y^2, x, y, 1\right) \), which is a set of six free coefficients \( A^T=\left\{A_{\text{xx}},A_{\text{xy}},A_{\text{yy}},A_x,A_y,A_0\right\} \). However, to specify an ellipse, all that is needed is five numbers; the major and minor axes lengths \( (a,b) \), the position \( (x_0,y_0) \), and the orientation \( \theta \). This is because the basis set includes lines, quadratics, parabolic and hyperbolic functions as well as elliptical functions as possible fits. If the fit is found to be a parabolic or hyperbolic function then the standard fitEllipse method is used. The AMS method restricts the fit to parabolic, hyperbolic and elliptical curves by imposing the condition that \( A^T ( D_x^T D_x + D_y^T D_y) A = 1 \) where the matrices \( Dx \) and \( Dy \) are the partial derivatives of the design matrix \( D \) with respect to x and y. The matrices are formed row by row applying the following to each of the points in the set:

\begin{align*} D(i,:)&=\left\{x_i^2, x_i y_i, y_i^2, x_i, y_i, 1\right\} & D_x(i,:)&=\left\{2 x_i,y_i,0,1,0,0\right\} & D_y(i,:)&=\left\{0,x_i,2 y_i,0,1,0\right\} \end{align*}

The AMS method minimizes the cost function

\begin{equation*} \epsilon ^2=\frac{ A^T D^T D A }{ A^T (D_x^T D_x + D_y^T D_y) A^T } \end{equation*}

The minimum cost is found by solving the generalized eigenvalue problem.

\begin{equation*} D^T D A = \lambda \left( D_x^T D_x + D_y^T D_y\right) A \end{equation*}

Parameters
pointsInput 2D point set, stored in std::vector<> or Mat
Examples:
fitellipse.cpp.

§ fitEllipseDirect()

RotatedRect cv::fitEllipseDirect ( InputArray  points)
Python:
retval=cv.fitEllipseDirect(points)

Fits an ellipse around a set of 2D points.

The function calculates the ellipse that fits a set of 2D points. It returns the rotated rectangle in which the ellipse is inscribed. The Direct least square (Direct) method by [60] is used.

For an ellipse, this basis set is \( \chi= \left(x^2, x y, y^2, x, y, 1\right) \), which is a set of six free coefficients \( A^T=\left\{A_{\text{xx}},A_{\text{xy}},A_{\text{yy}},A_x,A_y,A_0\right\} \). However, to specify an ellipse, all that is needed is five numbers; the major and minor axes lengths \( (a,b) \), the position \( (x_0,y_0) \), and the orientation \( \theta \). This is because the basis set includes lines, quadratics, parabolic and hyperbolic functions as well as elliptical functions as possible fits. The Direct method confines the fit to ellipses by ensuring that \( 4 A_{xx} A_{yy}- A_{xy}^2 > 0 \). The condition imposed is that \( 4 A_{xx} A_{yy}- A_{xy}^2=1 \) which satisfies the inequality and as the coefficients can be arbitrarily scaled is not overly restrictive.

\begin{equation*} \epsilon ^2= A^T D^T D A \quad \text{with} \quad A^T C A =1 \quad \text{and} \quad C=\left(\begin{matrix} 0 & 0 & 2 & 0 & 0 & 0 \\ 0 & -1 & 0 & 0 & 0 & 0 \\ 2 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 \end{matrix} \right) \end{equation*}

The minimum cost is found by solving the generalized eigenvalue problem.

\begin{equation*} D^T D A = \lambda \left( C\right) A \end{equation*}

The system produces only one positive eigenvalue \( \lambda\) which is chosen as the solution with its eigenvector \(\mathbf{u}\). These are used to find the coefficients

\begin{equation*} A = \sqrt{\frac{1}{\mathbf{u}^T C \mathbf{u}}} \mathbf{u} \end{equation*}

The scaling factor guarantees that \(A^T C A =1\).

Parameters
pointsInput 2D point set, stored in std::vector<> or Mat
Examples:
fitellipse.cpp.

§ fitLine()

void cv::fitLine ( InputArray  points,
OutputArray  line,
int  distType,
double  param,
double  reps,
double  aeps 
)
Python:
line=cv.fitLine(points, distType, param, reps, aeps[, line])

Fits a line to a 2D or 3D point set.

The function fitLine fits a line to a 2D or 3D point set by minimizing \(\sum_i \rho(r_i)\) where \(r_i\) is a distance between the \(i^{th}\) point, the line and \(\rho(r)\) is a distance function, one of the following:

  • DIST_L2

    \[\rho (r) = r^2/2 \quad \text{(the simplest and the fastest least-squares method)}\]

  • DIST_L1

    \[\rho (r) = r\]

  • DIST_L12

    \[\rho (r) = 2 \cdot ( \sqrt{1 + \frac{r^2}{2}} - 1)\]

  • DIST_FAIR

    \[\rho \left (r \right ) = C^2 \cdot \left ( \frac{r}{C} - \log{\left(1 + \frac{r}{C}\right)} \right ) \quad \text{where} \quad C=1.3998\]

  • DIST_WELSCH

    \[\rho \left (r \right ) = \frac{C^2}{2} \cdot \left ( 1 - \exp{\left(-\left(\frac{r}{C}\right)^2\right)} \right ) \quad \text{where} \quad C=2.9846\]

  • DIST_HUBER

    \[\rho (r) = \fork{r^2/2}{if \(r < C\)}{C \cdot (r-C/2)}{otherwise} \quad \text{where} \quad C=1.345\]

The algorithm is based on the M-estimator ( http://en.wikipedia.org/wiki/M-estimator ) technique that iteratively fits the line using the weighted least-squares algorithm. After each iteration the weights \(w_i\) are adjusted to be inversely proportional to \(\rho(r_i)\) .

Parameters
pointsInput vector of 2D or 3D points, stored in std::vector<> or Mat.
lineOutput line parameters. In case of 2D fitting, it should be a vector of 4 elements (like Vec4f) - (vx, vy, x0, y0), where (vx, vy) is a normalized vector collinear to the line and (x0, y0) is a point on the line. In case of 3D fitting, it should be a vector of 6 elements (like Vec6f) - (vx, vy, vz, x0, y0, z0), where (vx, vy, vz) is a normalized vector collinear to the line and (x0, y0, z0) is a point on the line.
distTypeDistance used by the M-estimator, see DistanceTypes
paramNumerical parameter ( C ) for some types of distances. If it is 0, an optimal value is chosen.
repsSufficient accuracy for the radius (distance between the coordinate origin and the line).
aepsSufficient accuracy for the angle. 0.01 would be a good default value for reps and aeps.

§ HuMoments() [1/2]

void cv::HuMoments ( const Moments moments,
double  hu[7] 
)
Python:
hu=cv.HuMoments(m[, hu])

Calculates seven Hu invariants.

The function calculates seven Hu invariants (introduced in [86]; see also http://en.wikipedia.org/wiki/Image_moment) defined as:

\[\begin{array}{l} hu[0]= \eta _{20}+ \eta _{02} \\ hu[1]=( \eta _{20}- \eta _{02})^{2}+4 \eta _{11}^{2} \\ hu[2]=( \eta _{30}-3 \eta _{12})^{2}+ (3 \eta _{21}- \eta _{03})^{2} \\ hu[3]=( \eta _{30}+ \eta _{12})^{2}+ ( \eta _{21}+ \eta _{03})^{2} \\ hu[4]=( \eta _{30}-3 \eta _{12})( \eta _{30}+ \eta _{12})[( \eta _{30}+ \eta _{12})^{2}-3( \eta _{21}+ \eta _{03})^{2}]+(3 \eta _{21}- \eta _{03})( \eta _{21}+ \eta _{03})[3( \eta _{30}+ \eta _{12})^{2}-( \eta _{21}+ \eta _{03})^{2}] \\ hu[5]=( \eta _{20}- \eta _{02})[( \eta _{30}+ \eta _{12})^{2}- ( \eta _{21}+ \eta _{03})^{2}]+4 \eta _{11}( \eta _{30}+ \eta _{12})( \eta _{21}+ \eta _{03}) \\ hu[6]=(3 \eta _{21}- \eta _{03})( \eta _{21}+ \eta _{03})[3( \eta _{30}+ \eta _{12})^{2}-( \eta _{21}+ \eta _{03})^{2}]-( \eta _{30}-3 \eta _{12})( \eta _{21}+ \eta _{03})[3( \eta _{30}+ \eta _{12})^{2}-( \eta _{21}+ \eta _{03})^{2}] \\ \end{array}\]

where \(\eta_{ji}\) stands for \(\texttt{Moments::nu}_{ji}\) .

These values are proved to be invariants to the image scale, rotation, and reflection except the seventh one, whose sign is changed by reflection. This invariance is proved with the assumption of infinite image resolution. In case of raster images, the computed Hu invariants for the original and transformed images are a bit different.

Parameters
momentsInput moments computed with moments .
huOutput Hu invariants.
See also
matchShapes

§ HuMoments() [2/2]

void cv::HuMoments ( const Moments m,
OutputArray  hu 
)
Python:
hu=cv.HuMoments(m[, hu])

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

§ intersectConvexConvex()

float cv::intersectConvexConvex ( InputArray  _p1,
InputArray  _p2,
OutputArray  _p12,
bool  handleNested = true 
)
Python:
retval, _p12=cv.intersectConvexConvex(_p1, _p2[, _p12[, handleNested]])

finds intersection of two convex polygons

§ isContourConvex()

bool cv::isContourConvex ( InputArray  contour)
Python:
retval=cv.isContourConvex(contour)

Tests a contour convexity.

The function tests whether the input contour is convex or not. The contour must be simple, that is, without self-intersections. Otherwise, the function output is undefined.

Parameters
contourInput vector of 2D points, stored in std::vector<> or Mat

§ matchShapes()

double cv::matchShapes ( InputArray  contour1,
InputArray  contour2,
int  method,
double  parameter 
)
Python:
retval=cv.matchShapes(contour1, contour2, method, parameter)

Compares two shapes.

The function compares two shapes. All three implemented methods use the Hu invariants (see HuMoments)

Parameters
contour1First contour or grayscale image.
contour2Second contour or grayscale image.
methodComparison method, see ShapeMatchModes
parameterMethod-specific parameter (not supported now).

§ minAreaRect()

RotatedRect cv::minAreaRect ( InputArray  points)
Python:
retval=cv.minAreaRect(points)

Finds a rotated rectangle of the minimum area enclosing the input 2D point set.

The function calculates and returns the minimum-area bounding rectangle (possibly rotated) for a specified point set. Developer should keep in mind that the returned RotatedRect can contain negative indices when data is close to the containing Mat element boundary.

Parameters
pointsInput vector of 2D points, stored in std::vector<> or Mat
Examples:
minarea.cpp.

§ minEnclosingCircle()

void cv::minEnclosingCircle ( InputArray  points,
Point2f center,
float &  radius 
)
Python:
center, radius=cv.minEnclosingCircle(points)

Finds a circle of the minimum area enclosing a 2D point set.

The function finds the minimal enclosing circle of a 2D point set using an iterative algorithm.

Parameters
pointsInput vector of 2D points, stored in std::vector<> or Mat
centerOutput center of the circle.
radiusOutput radius of the circle.
Examples:
minarea.cpp.

§ minEnclosingTriangle()

double cv::minEnclosingTriangle ( InputArray  points,
OutputArray  triangle 
)
Python:
retval, triangle=cv.minEnclosingTriangle(points[, triangle])

Finds a triangle of minimum area enclosing a 2D point set and returns its area.

The function finds a triangle of minimum area enclosing the given set of 2D points and returns its area. The output for a given 2D point set is shown in the image below. 2D points are depicted in red* and the enclosing triangle in yellow.

minenclosingtriangle.png
Sample output of the minimum enclosing triangle function

The implementation of the algorithm is based on O'Rourke's [143] and Klee and Laskowski's [97] papers. O'Rourke provides a \(\theta(n)\) algorithm for finding the minimal enclosing triangle of a 2D convex polygon with n vertices. Since the minEnclosingTriangle function takes a 2D point set as input an additional preprocessing step of computing the convex hull of the 2D point set is required. The complexity of the convexHull function is \(O(n log(n))\) which is higher than \(\theta(n)\). Thus the overall complexity of the function is \(O(n log(n))\).

Parameters
pointsInput vector of 2D points with depth CV_32S or CV_32F, stored in std::vector<> or Mat
triangleOutput vector of three 2D points defining the vertices of the triangle. The depth of the OutputArray must be CV_32F.
Examples:
minarea.cpp.

§ moments()

Moments cv::moments ( InputArray  array,
bool  binaryImage = false 
)
Python:
retval=cv.moments(array[, binaryImage])

Calculates all of the moments up to the third order of a polygon or rasterized shape.

The function computes moments, up to the 3rd order, of a vector shape or a rasterized shape. The results are returned in the structure cv::Moments.

Parameters
arrayRaster image (single-channel, 8-bit or floating-point 2D array) or an array ( \(1 \times N\) or \(N \times 1\) ) of 2D points (Point or Point2f ).
binaryImageIf it is true, all non-zero image pixels are treated as 1's. The parameter is used for images only.
Returns
moments.
Note
Only applicable to contour moments calculations from Python bindings: Note that the numpy type for the input array should be either np.int32 or np.float32.
See also
contourArea, arcLength

§ pointPolygonTest()

double cv::pointPolygonTest ( InputArray  contour,
Point2f  pt,
bool  measureDist 
)
Python:
retval=cv.pointPolygonTest(contour, pt, measureDist)

Performs a point-in-contour test.

The function determines whether the point is inside a contour, outside, or lies on an edge (or coincides with a vertex). It returns positive (inside), negative (outside), or zero (on an edge) value, correspondingly. When measureDist=false , the return value is +1, -1, and 0, respectively. Otherwise, the return value is a signed distance between the point and the nearest contour edge.

See below a sample output of the function where each image pixel is tested against the contour:

pointpolygon.png
sample output
Parameters
contourInput contour.
ptPoint tested against the contour.
measureDistIf true, the function estimates the signed distance from the point to the nearest contour edge. Otherwise, the function only checks if the point is inside a contour or not.

§ rotatedRectangleIntersection()

int cv::rotatedRectangleIntersection ( const RotatedRect rect1,
const RotatedRect rect2,
OutputArray  intersectingRegion 
)
Python:
retval, intersectingRegion=cv.rotatedRectangleIntersection(rect1, rect2[, intersectingRegion])

Finds out if there is any intersection between two rotated rectangles.

If there is then the vertices of the intersecting region are returned as well.

Below are some examples of intersection configurations. The hatched pattern indicates the intersecting region and the red vertices are returned by the function.

intersection.png
intersection examples
Parameters
rect1First rectangle
rect2Second rectangle
intersectingRegionThe output array of the vertices of the intersecting region. It returns at most 8 vertices. Stored as std::vector<cv::Point2f> or cv::Mat as Mx1 of type CV_32FC2.
Returns
One of RectanglesIntersectTypes