DLEllipseDetect Class Reference

#include <DLEllipseDetect.h>

List of all members.

Public Member Functions

 DLEllipseDetect ()
 DLEllipseDetect (DLImage &image, DL_BYTE mag_thres, double image_shrink_ratio, int area_min_pixel, int area_max_pixel, float axis_ratio_max, float y_center0=0, float y_center1=1.0, float x_center0=0, float x_center1=1.0, int ny_center=100, int nx_center=100, int narea=100)
virtual ~DLEllipseDetect ()
void dlDrawEllipseRegion (DLImage &outputImage, DL3DBin *bins, int binIndex, enum DLColorMapValue markerColor=DL_RED1)
list
< DLEllipseCandidate
dlGetDetectedEllipses ()


Constructor & Destructor Documentation

DLEllipseDetect::DLEllipseDetect (  )  [inline]

Default DLEllipseDetect Constructor

DLEllipseDetect::DLEllipseDetect ( DLImage &  image,
DL_BYTE  mag_thres,
double  image_shrink_ratio,
int  area_min_pixel,
int  area_max_pixel,
float  axis_ratio_max,
float  y_center0 = 0,
float  y_center1 = 1.0,
float  x_center0 = 0,
float  x_center1 = 1.0,
int  ny_center = 100,
int  nx_center = 100,
int  narea = 100 
)

DLEllipseDetect Constructor

Detects elliptic and complex shape objects from input images and return a list detected ellipses with their corresponding confidence values.

Parameters:
image DLImage input image
mag_thres DL_BYTE threshold value used for evaluating the edge magnitude image. Edges with magnitude weaker than this value are ignored. Its valid range is [1, 255]
image_shrink_ratio double the image shrink ratio (>=1)
A priori information of the wanted elliptic objects can be used to effectively improve the detection accuracy and further constraint the search space for faster processing. If currently there is no available a priori information on the area of the searched elliptic objects, you can simply input area_min_pixel=0, area_max_pixel=rows*cols. In this case, there are no constraints posed on the range of ellipse area.

Parameters:
area_min_pixel int the minimum area of the elliptic objects in pixel square in original input image
area_max_pixel int the maximum area of the elliptic objects in pixel square in original input image
axis_ratio_max float the maximum possible ratio of the lengths of the major axis to the minor axis
The following four parameters specify the region of the search within the image with the full range of each scaled to [0, 1.0]. x_left=0, x_right=1.0, y_left=0, y_right=1.0 specify that the search is to be performed in the entire image.
Parameters:
y_center0 float starting position in row, valid range [0, 1.0]
y_center1 float ending position in row, valid range [0, 1.0]
x_center0 float starting position in column, valid range [0, 1.0]
x_center1 float ending position in column, valid range [0, 1.0]
The following three parameters specify the number of accumulator bins for voting in each dimension, i.e. the center of the ellipse in row direction, the center of the ellipse in column direction, and the area of the ellipse.
Parameters:
ny_center int the number of distinctive accumulator bins for ellipse center in row direction
nx_center int the number of distinctive accumulator bins for ellipse center in column direction
narea int the number of distinctive accumulator bins for ellipse area

virtual DLEllipseDetect::~DLEllipseDetect (  )  [inline, virtual]

Default DLEllipseDetect Destructor


Member Function Documentation

void DLEllipseDetect::dlDrawEllipseRegion ( DLImage &  outputImage,
DL3DBin bins,
int  binIndex,
enum DLColorMapValue  markerColor = DL_RED1 
)

Draw colored bounding boxes around the specified ellipse candidates

Parameters:
outputImage DLImage output color image
bins DL3DBin* to 3-dimensional accumulator bins
binIndex int the index of the 3-dimensional accumulator bin
markerColor DLColorType the color of the plotted bounding box

list<DLEllipseCandidate> DLEllipseDetect::dlGetDetectedEllipses (  )  [inline]

Return the list of ellipse candidates

Returns:
the list of ellipse candidates in DLEllipseCandidate type


The documentation for this class was generated from the following files:

StampDetectDOCLIB Library is an add-on of DOCLIB. DOCLIB is being developed under contract by a collaboration between:
The Laboratory for Language and Media Processing
Unviersity of Maryland, College Park
and
Booz | Allen | Hamilton

All Rights Reserved, 2003-2007