geometric_voting_plugin Class Reference

Recognition-method geometric-voting. More...

#include <geometric_voting_plugin.hh>

Inheritance diagram for geometric_voting_plugin:

[legend]Collaboration diagram for geometric_voting_plugin:

[legend]List of all members.

Public Member Functions
	geometric_voting_plugin (const calibration_ptr &_calib, const mimas::mm_xml_reference< XalanElement > &configuration, bool _verbose)
	Constructor.
virtual std::vector< match >	recognise (const mimas::mm_image< unsigned char > &image, double timestamp, mi_scene_ptr scene) const
	Recognise objects using geometric voting.
Protected Attributes
int	fringe
	Width of boundary to cut from image.
int	sampling_rate
	Sampling rate for sub-sampling image.
double	timeout
	Timeout for recognition.
short int	min_gradient
	Gradient threshold for feature extraction.
std::map< std::string, voting_table_ptr >	models
	Map of voting tables.
mimas::mm_image_qtoutput< mimas::mm_rgba< unsigned char > >	display
	Display for debugging.

---

Detailed Description

Recognition-method geometric-voting.

See chapter about geometric hashing in the computer vision book from Prentice Hall:

    Computer Vision - A Modern Approach
    Prentice Hall 2002
    

Author:

Jan Wedekind (j.wedekind@shu.ac.uk)

---

Constructor & Destructor Documentation

geometric_voting_plugin::geometric_voting_plugin (  const calibration_ptr &  _calib, const mimas::mm_xml_reference< XalanElement > &  configuration, bool  _verbose )

Constructor. The constructor will extract the configuration from the XML-tree. For all micro-objects to be recognized with this method, voting_table objects are created. Parameters: _calib  Calibration data. configuration  Top-element of vision-configuration. _verbose  Set to true, if you want to get verbose information from the vision-algorithms.

---

Member Function Documentation

virtual std::vector< match > geometric_voting_plugin::recognise (  const mimas::mm_image< unsigned char > &  image, double  timestamp, mi_scene_ptr  scene )  const [virtual]

Recognise objects using geometric voting. Parameters: image  Image to recognise objects in. timestamp  Timestamp for image scene  Scene information. Implements recognition_method.

---

Member Data Documentation

int geometric_voting_plugin::fringe [protected]

Width of boundary to cut from image. Cutting of the boundary is required, when the image-source generates a black frame, which introduces strong gradient-features.

std::map< std::string, voting_table_ptr > geometric_voting_plugin::models [protected]

Map of voting tables. The name of a micro-object-type is mapped on a voting_table. See also: voting_table

int geometric_voting_plugin::sampling_rate [protected]

Sampling rate for sub-sampling image. For performance reasons the image can be sub-sampled before processing with the specified sampling-rate.

---

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

• geometric_voting_plugin.hh

---