MeSite¶

class MeSite(*args, **kwargs)¶

Bases: pybind11_object

Performs search in a given direction(normal) for a given distance(pixels) for a given ‘site’. Gives the most likely site given the probability from an ME mask.

Bug fix: rewrote application of masks to use the temporal information instead of applying both temporal masks to the same image. ie: spatial -> spatio/temporal
Added new tracking function to choose the most similar edge amongst all edges found.
sample step.

Overloaded function.

__init__(self: visp._visp.me.MeSite) -> None

Default constructor.

__init__(self: visp._visp.me.MeSite, ip: float, jp: float) -> None

Constructor from pixel coordinates.

__init__(self: visp._visp.me.MeSite, mesite: visp._visp.me.MeSite) -> None

Copy constructor.

Methods

`__init__`	Overloaded function.
`computeFinalThreshold`	Get the final computed likelihood threshold value, depending on the likelihood threshold type and ME settings.
`display`	Overloaded function.
`displayMeSite`	Overloaded function.
`distance`	Compute the square root distance between two moving-edges sites \(\|S1 - S2\| = \sqrt{(i_1-i_2)^2+(j_1-j_2)^2}\)
`getAlpha`	Get the angle of tangent at site.
`getContrastThreshold`	Get the contrast threshold of the site.
`getState`	Get the state of the site.
`getWeight`	Return site weight or uncertainty as a probability between 0 and 1.
`get_i`	Return integer coordinate along i (rows).
`get_ifloat`	Return subpixel coordinate along i (rows).
`get_j`	Return integer coordinate along j (columns).
`get_jfloat`	Return subpixel coordinate along j (columns).
`init`	Overloaded function.
`setAlpha`	Set the angle of tangent at site.
`setContrastThreshold`	Set the contrast threshold of the site.
`setDisplay`	Display selector.
`setState`
`setWeight`	Set the weight or uncertainty of the site.
`sqrDistance`	Compute the square distance between two moving-edges sites \(\|S1 - S2\| = (i_1-i_2)^2+(j_1-j_2)^2\)
`trackMultipleHypotheses`	Similar to the track() function, but stores the best numCandidates hypotheses in outputHypotheses .

Inherited Methods

Operators

`__doc__`
`__init__`	Overloaded function.
`__module__`
`__ne__`	Comparison operator.
`__repr__`

Attributes

`CONTRAST`
`M_ESTIMATOR`
`NONE`
`NO_SUPPRESSION`
`OUTSIDE_ROI_MASK`
`RANGE`
`RANGE_RESULT`
`RESULT`
`THRESHOLD`
`TOO_NEAR`
`UNKNOW`
`__annotations__`
`alpha`
`contrastThreshold`
`convlt`
`i`
`ifloat`
`j`
`jfloat`
`mask_sign`
`normGradient`
`weight`

class MeSiteDisplayType(self, value: int)¶

Bases: pybind11_object

Moving-edge site state

Values:

NO_SUPPRESSION: Point successfully tracked.
CONTRAST: Point not tracked due to a contrast problem, but retained in the ME list.
CONSTRAST: Deprecated. Point not tracked due to a likelihood problem, but retained in the ME list. Use instead CONTRAST.
THRESHOLD: Point not tracked due to the likelihood that is below the threshold, but retained in the ME list.
M_ESTIMATOR: Point detected as an outlier during virtual visual-servoing.
TOO_NEAR: Point not tracked anymore, since too near from its neighbor.
UNKNOW: Reserved.
OUTSIDE_ROI_MASK: Point is outside the region of interest mask, but retained in the ME list.

__and__(self, other: object) → object¶

__eq__(self, other: object) → bool¶

__ge__(self, other: object) → bool¶

__getstate__(self) → int¶

__gt__(self, other: object) → bool¶

__hash__(self) → int¶

__index__(self) → int¶

__init__(self, value: int)¶

__int__(self) → int¶

__invert__(self) → object¶

__le__(self, other: object) → bool¶

__lt__(self, other: object) → bool¶

__ne__(self, other: object) → bool¶

__or__(self, other: object) → object¶

__rand__(self, other: object) → object¶

__ror__(self, other: object) → object¶

__rxor__(self, other: object) → object¶

__setstate__(self, state: int) → None¶

__xor__(self, other: object) → object¶

property name : str¶

class MeSiteState(self, value: int)¶

Bases: pybind11_object

Moving-edge site state

Values:

NO_SUPPRESSION: Point successfully tracked.
CONTRAST: Point not tracked due to a contrast problem, but retained in the ME list.
CONSTRAST: Deprecated. Point not tracked due to a likelihood problem, but retained in the ME list. Use instead CONTRAST.
THRESHOLD: Point not tracked due to the likelihood that is below the threshold, but retained in the ME list.
M_ESTIMATOR: Point detected as an outlier during virtual visual-servoing.
TOO_NEAR: Point not tracked anymore, since too near from its neighbor.
UNKNOW: Reserved.
OUTSIDE_ROI_MASK: Point is outside the region of interest mask, but retained in the ME list.

__and__(self, other: object) → object¶

__eq__(self, other: object) → bool¶

__ge__(self, other: object) → bool¶

__getstate__(self) → int¶

__gt__(self, other: object) → bool¶

__hash__(self) → int¶

__index__(self) → int¶

__init__(self, value: int)¶

__int__(self) → int¶

__invert__(self) → object¶

__le__(self, other: object) → bool¶

__lt__(self, other: object) → bool¶

__ne__(self, other: object) → bool¶

__or__(self, other: object) → object¶

__rand__(self, other: object) → object¶

__ror__(self, other: object) → object¶

__rxor__(self, other: object) → object¶

__setstate__(self, state: int) → None¶

__xor__(self, other: object) → object¶

property name : str¶

__init__(*args, **kwargs)¶

Overloaded function.

__init__(self: visp._visp.me.MeSite) -> None

Default constructor.

__init__(self: visp._visp.me.MeSite, ip: float, jp: float) -> None

Constructor from pixel coordinates.

__init__(self: visp._visp.me.MeSite, mesite: visp._visp.me.MeSite) -> None

Copy constructor.

__ne__(self, m: visp.me.MeSite) → int¶: Comparison operator.

computeFinalThreshold(self, me: visp.me.Me) → float¶

Get the final computed likelihood threshold value, depending on the likelihood threshold type and ME settings.

Returns:: value of the contrast threshold of the site.

display(*args, **kwargs)¶

Overloaded function.

display(self: visp._visp.me.MeSite, I: visp._visp.core.ImageGray) -> None

Display moving edges in image I.

Parameters:

I: Input image.

display(self: visp._visp.me.MeSite, I: visp._visp.core.ImageRGBa) -> None

Display moving edges in image I.

Parameters:

I: Input image.

static displayMeSite(*args, **kwargs)¶

Overloaded function.

displayMeSite(I: visp._visp.core.ImageGray, i: float, j: float, state: visp._visp.me.MeSite.MeSiteState = NO_SUPPRESSION) -> None
displayMeSite(I: visp._visp.core.ImageRGBa, i: float, j: float, state: visp._visp.me.MeSite.MeSiteState = NO_SUPPRESSION) -> None

static distance(S1: visp.me.MeSite, S2: visp.me.MeSite) → float¶

Compute the square root distance between two moving-edges sites \(|S1 - S2| = \sqrt{(i_1-i_2)^2+(j_1-j_2)^2}\)

Note

See sqrDistance()

Parameters:

S1: visp.me.MeSite¶: First site
S2: visp.me.MeSite¶: Second site

Returns:

the distance between the two sites.

getAlpha(self) → float¶

Get the angle of tangent at site.

Returns:: value of alpha

getContrastThreshold(self) → float¶

Get the contrast threshold of the site.

Returns:: value of the contrast threshold of the site.

getState(self) → visp.me.MeSite.MeSiteState¶

Get the state of the site.

Returns:: flag corresponding to vpMeSiteState

getWeight(self) → float¶: Return site weight or uncertainty as a probability between 0 and 1.

get_i(self) → int¶: Return integer coordinate along i (rows).

Note

See get_ifloat()

get_ifloat(self) → float¶: Return subpixel coordinate along i (rows).

Note

See get_i()

get_j(self) → int¶: Return integer coordinate along j (columns).

Note

See get_jfloat()

get_jfloat(self) → float¶: Return subpixel coordinate along j (columns).

Note

See get_j()

init(*args, **kwargs)¶

Overloaded function.

init(self: visp._visp.me.MeSite) -> None

Initialize moving-edge site with default parameters.

init(self: visp._visp.me.MeSite, ip: float, jp: float, alphap: float) -> None

Initialize moving-edge site parameters.

init(self: visp._visp.me.MeSite, ip: float, jp: float, alphap: float, convltp: float) -> None

Initialize moving-edge site parameters.

init(self: visp._visp.me.MeSite, ip: float, jp: float, alphap: float, convltp: float, sign: int) -> None

Initialize moving-edge site parameters.

init(self: visp._visp.me.MeSite, ip: float, jp: float, alphap: float, convltp: float, sign: int, contrastThreshold: float) -> None

Initialize moving-edge site parameters.

setAlpha(self, a: float) → None¶

Set the angle of tangent at site.

Parameters:

a: float¶: new value of alpha

setContrastThreshold(self, thresh: float, me: visp.me.Me) → None¶

Set the contrast threshold of the site. If the vpMe::m_useAutomaticThreshold is set to false, the contrast threshold is set to the global value retrieved using vpMe::getThreshold() . This value can be set using vpMe::setThreshold() . Otherwise, the contrast threshold will be set to the highest value between thresh and the minimum value set by vpMe::setMinThreshold() that could be retrieved using vpMe::getMinThreshold() .

Parameters:

thresh: float¶: new value of contrast threshold
me: visp.me.Me¶: moving-edge parameters

setDisplay(self, select: visp.me.MeSite.MeSiteDisplayType) → None¶: Display selector.

setState(self, flag: visp.me.MeSite.MeSiteState) → None¶

setWeight(self, weight: float) → None¶

Set the weight or uncertainty of the site.

Parameters:

weight: float¶: New value of weight as a probability between 0 and 1.

static sqrDistance(S1: visp.me.MeSite, S2: visp.me.MeSite) → float¶

Compute the square distance between two moving-edges sites \(|S1 - S2| = (i_1-i_2)^2+(j_1-j_2)^2\)

Note

See distance()

Parameters:

S1: visp.me.MeSite¶: First site
S2: visp.me.MeSite¶: Second site

Returns:

The square distance between the two sites.

trackMultipleHypotheses(self, I: visp.core.ImageGray, me: visp.me.Me, test_contrast: bool, outputHypotheses: list[visp.me.MeSite], numCandidates: int) → list[visp.me.MeSite]¶

Similar to the track() function, but stores the best numCandidates hypotheses in outputHypotheses . The best matching hypotheses (if it is not suppressed) is assigned to this and is stored as the first element of outputHypotheses . The hypotheses are sorted from best to worst match in the vector. A match may be in the vector but mark as suppressed. If this is undesired, you should filter them afterwards.

Warning

To display the moving edges graphics a call to vpDisplay::flush() is needed after this function.

Returns:

A tuple containing:

outputHypotheses