centroidMulti Class Reference

Inheritance diagram for centroidMulti:

Collaboration diagram for centroidMulti:

Public Member Functions
def	__init__ (self, iPt=None, params=CfgCentMulti())
	Centroid track-pointer constructor. More...
def	get (self, fname)
	Get parameter of the tracker. More...
def	measure (self, I)
	Measure the track point from the given image. More...
def	process (self, I)
	Process the input image according to centroid tracking. More...
def	regionProposal (I)
	Find out the centroid for multiple objects. More...
def	set (self, fname, fval)
	Set parameters for the tracker. More...
Public Member Functions inherited from centroid
def	adapt (self)
	Adapt internal tracking parameters. More...
def	correct (self)
	Correct track state based on predication and measurement. More...
def	display_cv (self, I, ratio=None, window_name="track point ")
	Use opencv display routines to plot the trackpoint along with the given image. More...
def	displayDebugState (self, dbstate=None)
	Displays internally stored intermediate process output. More...
def	displayState (self, dstate=None, ax=None)
	Displays the current track pointer measurement. More...
def	emptyState (self)
	Return an empty state structure (in python a dataclass). More...
def	getState (self)
	Return the track-pointer state. More...
def	offset (self, dp)
	Apply a vector offset to the track point. More...
def	predict (self)
	Predict next track state based on current track state. More...
def	setIfMissing (self, params, pname, pval)
	Set missing parameters in the registration parameters structure. More...
def	setState (self, dPt)
	Use a track state element to define internal track state. More...
def	transform (self, g)
	Apply a Lie group transformation (linear/affine) to the track point. More...

Public Attributes
	haveMeas
	labelIm
	labelImage
	tpt
	trackProps
Public Attributes inherited from centroid
	haveMeas
	params
	tpt

Detailed Description

@ingroup  TrackPointer
@brief    Used to track the centroid of multiple (binarized) objects.
 
Performs tracking of multiple objects by computing their centroids.
 
There are two ways to perform centroid tracking.  One is to presume
that the object to track has already been binarized and it is simply a
matter of computing the centroid.  The other is that the binarization
needs to first be computed, then the centroid calculated.  If
binarization of the input is desired, use an ``improcessor`` to perform the
binarization as a pre-processing step.  If the binarization is too
complicated for an improcess procedure, then write a tracker wrapper around
the binarization + tracking combination. It will usually then become a
``perceiver`` object.

Constructor & Destructor Documentation

__init__()

def __init__	(		self,
			iPt = None,
			params = CfgCentMulti()
	)

Centroid track-pointer constructor.

Parameters

[in]	params	Parameter settings.
[in]	iPt	Initial track point coordinates.

Todo:

Should swap order if initial point is optional but params have non-trivial definition if not given. Really, should have a None check instead. Why was this particular choice made? Seems to lack flexibility or is it related to best practice?

Reimplemented from centroid.

Member Function Documentation

get()

def get	(		self,
			fname
	)

Get parameter of the tracker.

Parameters

[in]	fname	Name of parameter.
[out]	fval	Value of parameter.

Todo:

Not coded, just the stub.

Reimplemented from centroid.

measure()

def measure	(		self,
			I
	)

Measure the track point from the given image.

Parameters

[in]	I	The input image.
[out]	mstate	Measured state, if desired as output.

Reimplemented from centroid.

process()

def process	(		self,
			I
	)

Process the input image according to centroid tracking.

Parameters

[in]

I

The input image.

Reimplemented from centroid.

regionProposal()

def regionProposal

(

I

)

Find out the centroid for multiple objects.

Parameters

[in]

I

The input mask image.

Note

Code here is horrendous. What is purpose here? There are better methods. Even the stackoverflow link in code (below) has one. 08/03 PAV.

Original code can be implemented with scikit labels. Note that regionprops assumes different areas are with different labels. See https://stackoverflow.com/a/61591279/5269146

set()

def set	(		self,
			fname,
			fval
	)

Set parameters for the tracker.

Parameters

[in]

fname

Name of parameter. fval Value of parameter.

Todo:

Not coded, just the stub.

Reimplemented from centroid.

Member Data Documentation

haveMeas

haveMeas

labelIm

labelIm

labelImage

labelImage

tpt

tpt

trackProps

trackProps

---

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

• /home/mary/python/trackpointer/trackpointer/centroidMulti.py