Matcher Class Reference

Generic puzzle piece matching class. More...

Inheritance diagram for Matcher:

Public Member Functions
def	__init__ (self, theParams=CfgMatcher)
	Constructor for the matcher class. More...
def	compare (self, piece_A, piece_B)
	Compare between two passed puzzle piece data. More...
tuple[float, np.ndarray]	estimateAffineMatch (self, piece_A, piece_B)
	Estimate the rigid affine transform that aligns piece A to piece B. More...
def	extractFeature (self, piece)
	Process raw puzzle piece data to obtain encoded description of piece. More...
"Matcher"	fit (self, groups, list[str]|None labels=None)
	Given raw data regarding expected element instances, identify model to differentiate them if possible for this matcher type. More...
"Matcher"	fitFromImageSegmented (self, Irgb, Iseg, hasLabs=False)
	Given an image and a segmentation, generate fit. More...
def	score (self, piece_A, piece_B)
	Compute the score between two passed puzzle piece data. More...
def	solveMatchedPuzzle (self, puzzle, sol)

Static Public Member Functions
tuple[np.ndarray, np.ndarray]	pcaFrame (piece)
	Estimate a signed PCA frame from a puzzle piece's foreground pixels. More...

Public Attributes
	params

Detailed Description

Generic puzzle piece matching class.

Actual instances should use similarity of difference matching sub-classes.

Constructor & Destructor Documentation

__init__()

def __init__	(		self,
			theParams = CfgMatcher
	)

Constructor for the matcher class.

Parameters

[in]

theParams

The matcher configuration (optional).

Reimplemented in MatchSimilar, SIFTCV, PCA, Moments, HistogramCV, Distance, and MatchDifferent.

Member Function Documentation

compare()

def compare	(		self,
			piece_A,
			piece_B
	)

Compare between two passed puzzle piece data.

This member function should be overloaded.

Parameters

[in]	piece_A	Puzzle piece A instance.
[in]	piece_B	Puzzle piece B instance.
[out]	Outcome	of matching classification, when function overloaded.

Reimplemented in Sift, MatchSimilar, and MatchDifferent.

estimateAffineMatch()

tuple[float, np.ndarray] estimateAffineMatch	(		self,
			piece_A,
			piece_B
	)

Estimate the rigid affine transform that aligns piece A to piece B.

The transform maps global pixel coordinates from piece_A into the global coordinate frame of piece_B. Its rotation is determined by their signed PCA frames and its translation maps the A centroid onto the B centroid.

Returns

(rotation_degrees, affine) where affine is a 3x3 homogeneous rigid transform.

extractFeature()

def extractFeature	(		self,
			piece
	)

Process raw puzzle piece data to obtain encoded description of piece.

Use to recognize/associate the piece given new measurements. This member function should be overloaded.

Parameters

[in]	piece	Template instance saving a piece's info.
[out]	featVec	The "feature" vector.

Reimplemented in SIFTCV, PCA, Moments, HistogramCV, Distance, ColorBoWMatcher, and Edge.

fit()

"Matcher" fit	(		self,
			groups,
		list[str] | None	labels = None
	)

Given raw data regarding expected element instances, identify model to differentiate them if possible for this matcher type.

This base class does nothing. Overload as fitting.

Parameters

groups	List of (3, N) RGB matrices forming the database.
labels	Optional list of human-readable names, one per group. Auto-generated as "group_0", "group_1", ... if None.

Returns

Self, to allow method chaining (e.g., matcher.fit(groups).query(q)).

fitFromImageSegmented()

"Matcher" fitFromImageSegmented	(		self,
			Irgb,
			Iseg,
			hasLabs = False
	)

Given an image and a segmentation, generate fit.

This base class does nothing. Overload as fitting.

Permits fitting based on two images, one with all objects of interest in it as a color image and the second a segmentation isolating the objects. It should be considered binary in nature, then no labels assumed. If it has labels (each unique value is the label), then snag from Iseg as labels.

Parameters

[in]	Irgb	Source color image.
[in]	Iseg	Binary segmentation of image, or label segmentation.
[in]	hasLabs	Iseg has labels and is interpreted as binary. Default: False.

Reimplemented in ColorBoWMatcher.

pcaFrame()

tuple[np.ndarray, np.ndarray] pcaFrame (   piece )

static

Estimate a signed PCA frame from a puzzle piece's foreground pixels.

Parameters

[in]

piece

Template puzzle piece.

Returns

(center, frame) where center is the global pixel centroid and frame is a 2x2 right-handed rotation matrix whose first column is the major PCA axis.

score()

def score	(		self,
			piece_A,
			piece_B
	)

Compute the score between two passed puzzle piece data.

Parameters

[in]	piece_A	Template instance saving a piece's info.
[in]	piece_B	Template instance saving a piece's info.
[out]	Distance	of the feature vectors. (Overload if not proper).

Reimplemented in Sift, SIFTCV, PCA, Moments, HistogramCV, and ColorBoWMatcher.

solveMatchedPuzzle()

def solveMatchedPuzzle	(		self,
			puzzle,
			sol
	)

Reimplemented in SIFTCV.

Member Data Documentation

params

params

---

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

• /home/mary/python/puzzle_solver/puzzle/pieces/matcher.py