Class containing functions and data relating to annotating outlines

Description

An AnnotatedOutline contains a function to annotate tears on the outline.

Value

AnnotatedOutline object, with extra fields for tears latitude of rim phi0 and index of fixed point i0.

Super classes

retistruct::OutlineCommon -> retistruct::Outline -> retistruct::PathOutline -> AnnotatedOutline

Public fields

Methods

Public methods

• AnnotatedOutline$new()

• AnnotatedOutline$labelTearPoints()

• AnnotatedOutline$whichTear()

• AnnotatedOutline$getTear()

• AnnotatedOutline$getTears()

• AnnotatedOutline$computeTearRelationships()

• AnnotatedOutline$addTear()

• AnnotatedOutline$removeTear()

• AnnotatedOutline$checkTears()

• AnnotatedOutline$setFixedPoint()

• AnnotatedOutline$getFixedPoint()

• AnnotatedOutline$getRimSet()

• AnnotatedOutline$getBoundarySets()

• AnnotatedOutline$ensureFixedPointInRim()

• AnnotatedOutline$labelFullCutPoints()

• AnnotatedOutline$addFullCut()

• AnnotatedOutline$whichFullCut()

• AnnotatedOutline$removeFullCut()

• AnnotatedOutline$computeFullCutRelationships()

• AnnotatedOutline$getFullCut()

• AnnotatedOutline$getFullCuts()

• AnnotatedOutline$addPoints()

• AnnotatedOutline$getRimLengths()

• AnnotatedOutline$clone()

Method new()

Constructor

Usage

AnnotatedOutline$new(...)

Arguments

Method labelTearPoints()

Label a set of three unlabelled points supposed to refer to the apex and vertices of a tear with the V0 (Apex), VF (forward vertex) and VB (backward vertex) labels.

Usage

AnnotatedOutline$labelTearPoints(pids)

Arguments

Returns

Vector of indices labelled with V0, VF and VB

Method whichTear()

Return index of tear in an AnnotatedOutline in which a point appears

Usage

AnnotatedOutline$whichTear(pid)

Arguments

Returns

ID of tear

Method getTear()

Return indices of tear in AnnotatedOutline

Usage

AnnotatedOutline$getTear(tid)

Arguments

Returns

Vector of three point IDs, labelled with V0, VF and VB

Method getTears()

Get tears

Usage

AnnotatedOutline$getTears()

Returns

Matrix of tears

Method computeTearRelationships()

Compute the parent relationships for a potential set of tears. The function throws an error if tears overlap.

Usage

AnnotatedOutline$computeTearRelationships(tears = NULL)

Arguments

Returns

List containing

Rset

the set of points on the rim

TFset

list containing indices of points in each forward tear

TBset

list containing indices of points in each backward tear

h

correspondence mapping

hf

correspondence mapping in forward direction for points on boundary

hb

correspondence mapping in backward direction for points on boundary

Method addTear()

Add tear to an AnnotatedOutline

Usage

AnnotatedOutline$addTear(pids)

Arguments

Method removeTear()

Remove tear from an AnnotatedOutline

Usage

AnnotatedOutline$removeTear(tid)

Arguments

Method checkTears()

Check that all tears are correct.

Usage

AnnotatedOutline$checkTears()

Returns

If all is OK, returns empty vector. If not, returns indices of problematic tears.

Method setFixedPoint()

Set fixed point

Usage

AnnotatedOutline$setFixedPoint(i0, name)

Arguments

Method getFixedPoint()

Get point ID of fixed point

Usage

AnnotatedOutline$getFixedPoint()

Returns

Point ID of fixed point

Method getRimSet()

Get point IDs of points on rim

Usage

AnnotatedOutline$getRimSet()

Returns

Point IDs of points on rim. If the outline has been stitched (see StitchedOutline), the point IDs will be ordered in the direction of the forward pointer.

Method getBoundarySets()

Get point IDs of points on boundaries

Usage

AnnotatedOutline$getBoundarySets()

Returns

List of Point IDs of points on the boundaries. If the outline has been stitched (see StitchedOutline), the point IDs in each element of the list will be ordered in the direction of the forward pointer, and the boundary that is longest will be named as Rim. If the outline has not been stitched, the list will have one element named Rim.

Method ensureFixedPointInRim()

Ensure that the fixed point i0 is in the rim, not a tear. Alters object in which i0 may have been changed.

Usage

AnnotatedOutline$ensureFixedPointInRim()

Method labelFullCutPoints()

Label a set of four unlabelled points supposed to refer to a cut.

Usage

AnnotatedOutline$labelFullCutPoints(pids)

Arguments

Method addFullCut()

Add cut to an AnnotatedOutline

Usage

AnnotatedOutline$addFullCut(pids)

Arguments

Method whichFullCut()

Return index of cut in an AnnotatedOutline in which a point appears

Usage

AnnotatedOutline$whichFullCut(pid)

Arguments

Returns

ID of cut

Method removeFullCut()

Remove cut from an AnnotatedOutline

Usage

AnnotatedOutline$removeFullCut(cid)

Arguments

Method computeFullCutRelationships()

Compute the cut relationships between the points

Usage

AnnotatedOutline$computeFullCutRelationships(fullcuts)

Arguments

Returns

List containing

Rset

the set of points on the rim

TFset

list containing indices of points in each forward cut

TBset

list containing indices of points in each backward cut

h

correspondence mapping

hf

correspondence mapping in forward direction for points on boundary

hb

correspondence mapping in backward direction for points on boundary

Method getFullCut()

Return indices of fullcuts in AnnotatedOutline

Usage

AnnotatedOutline$getFullCut(cid)

Arguments

Returns

Vector of four point IDs, labelled with VF1, VF1, VB0 and VB1

Method getFullCuts()

Return indices of fullcuts in AnnotatedOutline

Usage

AnnotatedOutline$getFullCuts()

Returns

Matrix in which each row contains point IDs, for the forward and backward sides of the cut: VF0, VF1, VB0 and VB1

Method addPoints()

Add points to the outline register of points

Usage

AnnotatedOutline$addPoints(P, fid)

Arguments

Returns

The ID of each added point in the register. If points already exist a point will not be created in the register, but an ID will be returned

Method getRimLengths()

Get lengths of edges on rim

Usage

AnnotatedOutline$getRimLengths()

Returns

Vector of rim lengths

Method clone()

The objects of this class are cloneable with this method.

Usage

AnnotatedOutline$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Examples

P <- rbind(c(1,1),   c(2,1),  c(2,-1),
           c(1,-1),  c(1,-2), c(-1,-2),
           c(-1,-1), c(-2,-1),c(-2,1),
           c(-1,1),  c(-1,2), c(1,2))
o <- TriangulatedOutline$new(P)
o$addTear(c(3, 4, 5))
o$addTear(c(6, 7, 8))
o$addTear(c(9, 10, 11))
o$addTear(c(12, 1, 2))
flatplot(o)

P <- list(rbind(c(1,1), c(2,1), c(2.5,2), c(3,1), c(4,1), c(1,4)),
          rbind(c(-1,1), c(-1,4), c(-2,3), c(-2,2), c(-3,2), c(-4,1)),
          rbind(c(-4,-1), c(-1,-1), c(-1,-4)),
          rbind(c(1,-1), c(2,-1), c(2.5,-2), c(3,-1), c(4,-1), c(1,-4)))
o <- AnnotatedOutline$new(P)
o$addTear(c(2, 3, 4))
o$addTear(c(17, 18, 19))
o$addTear(c(9, 10, 11))
o$addFullCut(c(1, 5, 16, 20))
flatplot(o)

Subclass of FeatureSet to represent counts centred on points

Description

A CountSet contains information about points located on Outlines. Each CountSet contains a list of matrices, each of which has columns labelled X and Y describing the cartesian coordinates (in the unscaled coordinate frame) of the centres of boxes in the Outline, and a column C representing the counts in those boxes.

Super classes

retistruct::FeatureSetCommon -> retistruct::FeatureSet -> CountSet

Methods

Public methods

• CountSet$new()

• CountSet$reconstruct()

• CountSet$clone()

Method new()

Constructor

Usage

CountSet$new(data = NULL, cols = NULL)

Arguments

Method reconstruct()

Map the CountSet to a ReconstructedOutline

Usage

CountSet$reconstruct(ro)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

CountSet$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

The deformation energy function

Description

The function that computes the energy (or error) of the deformation of the mesh from the flat outline to the sphere. This depends on the locations of the points given in spherical coordinates. The function is designed to take these as a vector that is received from the optim function.

Usage

E(
  p,
  Cu,
  C,
  L,
  B,
  Tr,
  A,
  R,
  Rset,
  i0,
  phi0,
  lambda0,
  Nphi,
  N,
  alpha = 1,
  x0,
  nu = 1,
  verbose = FALSE
)

Arguments

Value

A single value, representing the energy of this particular configuration

Author(s)

David Sterratt

The deformation energy function

Description

The function that computes the energy (or error) of the deformation of the mesh from the flat outline to the sphere. This depends on the locations of the points given in spherical coordinates. The function is designed to take these as a vector that is received from the optim function.

Usage

Ecart(P, Cu, L, Tr, A, R, alpha = 1, x0, nu = 1, verbose = FALSE)

Arguments

Value

A single value, representing the energy of this particular configuration

Author(s)

David Sterratt

The deformation energy gradient function

Description

The function that computes the gradient of the energy (or error) of the deformation of the mesh from the flat outline to the sphere. This depends on the locations of the points given in spherical coordinates. The function is designed to take these as a vector that is received from the optim function.

Usage

Fcart(P, C, L, Tr, A, R, alpha = 1, x0, nu = 1, verbose = FALSE)

Arguments

Value

A vector representing the derivative of the energy of this particular configuration with respect to the parameter vector

Author(s)

David Sterratt

Superclass containing functions and data relating to sets of features in flat Outlines

Description

A FeatureSet contains information about features located on Outlines. Each FeatureSet contains a list of matrices, each of which has columns labelled X and Y describing the cartesian coordinates of points on the Outline, in the unscaled coordinate frame. Derived classes, e.g. a CountSet, may have extra columns. Each matrix in the list has an associated label and colour, which is used by plotting functions.

Super class

retistruct::FeatureSetCommon -> FeatureSet

Methods

Public methods

• FeatureSet$new()

• FeatureSet$clone()

Method new()

Constructor

Usage

FeatureSet$new(data = NULL, cols = NULL, type = NULL)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

FeatureSet$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functionality common to FeatureSets and ReconstructedFeatureSets

Description

An FeatureSetCommon has functionality for retrieving sets of features (e.g. points or landmarks associated with an outline)

Public fields

Methods

Public methods

• FeatureSetCommon$getIndex()

• FeatureSetCommon$getIDs()

• FeatureSetCommon$setID()

• FeatureSetCommon$getFeature()

• FeatureSetCommon$getFeatures()

• FeatureSetCommon$getCol()

• FeatureSetCommon$clone()

Method getIndex()

Get numeric index of features

Usage

FeatureSetCommon$getIndex(fid)

Arguments

Method getIDs()

Get IDs of features

Usage

FeatureSetCommon$getIDs()

Returns

Vector of IDs of features

Method setID()

Set name

Usage

FeatureSetCommon$setID(i, fid)

Arguments

Method getFeature()

Get feature by feature ID

Usage

FeatureSetCommon$getFeature(fid)

Arguments

Returns

Matrix describing feature

Method getFeatures()

Get all features

Usage

FeatureSetCommon$getFeatures()

Method getCol()

Get colour in which to plot feature ID

Usage

FeatureSetCommon$getCol(fid)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

FeatureSetCommon$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Construct an outline object. This sanitises the input points P, as described below.

Description

Construct an outline object. This sanitises the input points P, as described below.

Construct an outline object. This sanitises the input points P, as described below.

Public fields

Methods

Public methods

• Fragment$initializeFromPoints()

• Fragment$clone()

Method initializeFromPoints()

Initialise a Fragment from a set of points

Usage

Fragment$initializeFromPoints(P)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

Fragment$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Subclass of FeatureSet to represent points

Description

A LandmarkSet contains information about points located on Outlines. Each LandmarkSet contains a list of matrices, each of which has columns labelled X and Y describing the cartesian coordinates (in the unscaled coordinate frame) of points in landmarks on the Outline.

Super classes

retistruct::FeatureSetCommon -> retistruct::FeatureSet -> LandmarkSet

Methods

Public methods

• LandmarkSet$new()

• LandmarkSet$reconstruct()

• LandmarkSet$clone()

Method new()

Constructor

Usage

LandmarkSet$new(data = NULL, cols = NULL)

Arguments

Method reconstruct()

Map the LandmarkSet to a ReconstructedOutline

Usage

LandmarkSet$reconstruct(ro)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

LandmarkSet$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing basic information about flat outlines

Description

An Outline has contains the polygon describing the outline and an image associated with the outline.

Super class

retistruct::OutlineCommon -> Outline

Public fields

Methods

Public methods

• Outline$new()

• Outline$getImage()

• Outline$replaceImage()

• Outline$mapFragment()

• Outline$mapPids()

• Outline$getDepth()

• Outline$addPoints()

• Outline$getFragmentPointIDs()

• Outline$getFragmentPoints()

• Outline$getFragment()

• Outline$getFragmentIDsFromPointIDs()

• Outline$getFragmentIDs()

• Outline$getPoints()

• Outline$getPointsXY()

• Outline$getPointsScaled()

• Outline$getRimSet()

• Outline$getOutlineSet()

• Outline$getOutlineLengths()

• Outline$addFeatureSet()

• Outline$clone()

Method new()

Construct an outline object. This sanitises the input points P.

Usage

Outline$new(
  fragments = list(),
  scale = NA,
  im = NULL,
  scalez = NA,
  dm = NULL,
  units = NA
)

Arguments

Method getImage()

Image accessor

Usage

Outline$getImage()

Returns

An image as a raster object

Method replaceImage()

Image setter

Usage

Outline$replaceImage(im)

Arguments

Method mapFragment()

Map the point IDs of a Fragment on the point IDs of this Outline

Usage

Outline$mapFragment(fragment, pids)

Arguments

Method mapPids()

Map references to points

Usage

Outline$mapPids(x, y, pids)

Arguments

Returns

New references to point indices in target

Method getDepth()

Get depth of points P

Usage

Outline$getDepth(P)

Arguments

Returns

Vector of unscaled Z coordinates of points P

Method addPoints()

Add points to the outline register of points

Usage

Outline$addPoints(P, fid)

Arguments

Returns

The ID of each added point in the register. If points already exist a point will not be created in the register, but an ID will be returned

Method getFragmentPointIDs()

Get the point IDs in a fragment

Usage

Outline$getFragmentPointIDs(fid)

Arguments

Returns

Vector of point IDs, i.e. indices of the rows in the matrices returned by getPoints and getPointsScaled

Method getFragmentPoints()

Get the points in a fragment

Usage

Outline$getFragmentPoints(fid)

Arguments

Returns

Vector of points

Method getFragment()

Get fragment

Usage

Outline$getFragment(fid)

Arguments

Returns

The Fragment object with ID fid

Method getFragmentIDsFromPointIDs()

Get fragment IDs from point IDS

Usage

Outline$getFragmentIDsFromPointIDs(pids)

Arguments

Returns

The Fragment ID to which each point belongs

Method getFragmentIDs()

Get fragment IDs

Usage

Outline$getFragmentIDs()

Returns

IDs of all fragments

Method getPoints()

Get unscaled mesh points

Usage

Outline$getPoints(pids = NULL)

Arguments

Returns

Matrix with columns X, Y and Z

Method getPointsXY()

Get X-Y coordinates of unscaled mesh points

Usage

Outline$getPointsXY(pids = NULL)

Arguments

Returns

Matrix with columns X and Y

Method getPointsScaled()

Get scaled mesh points

Usage

Outline$getPointsScaled()

Returns

Matrix with columns X and Y which is exactly scale times the matrix returned by getPoints

Method getRimSet()

Get set of points on rim

Usage

Outline$getRimSet()

Returns

Vector of point IDs, i.e. indices of the rows in the matrices returned by getPoints and getPointsScaled

Method getOutlineSet()

Get points on the edge of the outline

Usage

Outline$getOutlineSet()

Returns

Vector of points IDs on outline

Method getOutlineLengths()

Get lengths of edges of the outline

Usage

Outline$getOutlineLengths()

Returns

Vector of lengths of edges connecting neighbouring points

Method addFeatureSet()

Add a FeatureSet, e.g. a PointSet or LandmarkSet

Usage

Outline$addFeatureSet(fs)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

Outline$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functionality common to flat and reconstructed outlines

Description

An OutlineCommon has functionality for retrieving sets of features (e.g. points or landmarks associated with an outline)

Public fields

Methods

Public methods

• OutlineCommon$getFeatureSets()

• OutlineCommon$getFeatureSet()

• OutlineCommon$clearFeatureSets()

• OutlineCommon$getIDs()

• OutlineCommon$getFeatureSetTypes()

• OutlineCommon$clone()

Method getFeatureSets()

Get all the feature sets

Usage

OutlineCommon$getFeatureSets()

Returns

List of FeatureSets associated with the outline

Method getFeatureSet()

Get all feature sets of a particular type, e.g. PointSet or LandmarkSet

Usage

OutlineCommon$getFeatureSet(type)

Arguments

Returns

All FeatureSets of that type

Method clearFeatureSets()

Clear all feature sets from the outline

Usage

OutlineCommon$clearFeatureSets()

Method getIDs()

Get all the distinct IDs contained in the FeatureSets

Usage

OutlineCommon$getIDs()

Returns

Vector of IDs

Method getFeatureSetTypes()

Get all the distinct types of FeatureSets

Usage

OutlineCommon$getFeatureSetTypes()

Returns

Vector of types as strings, e.g. PointSet, LandmarkSet

Method clone()

The objects of this class are cloneable with this method.

Usage

OutlineCommon$clone(deep = FALSE)

Arguments

Add point fullcuts to the outline

Description

Add point fullcuts to the outline

Add point fullcuts to the outline

Details

The member function stitchSubpaths() stitches together two subpaths of the outline. One subpath is stitched in the forward direction from the point indexed by VF0 to the point indexed by VF1. The other is stitched in the backward direction from VB0 to VB1. Each point in the subpath is linked to points in the opposing pathway at an equal or near-equal fraction along. If a point exists in the opposing pathway within a distance epsilon of the projection, this point is connected. If no point exists within this tolerance, a new point is created.

Value

To the Outline object this adds

Super classes

retistruct::OutlineCommon -> retistruct::Outline -> PathOutline

Public fields

Methods

Public methods

• PathOutline$addPoints()

• PathOutline$nextPoint()

• PathOutline$insertPoint()

• PathOutline$stitchSubpaths()

• PathOutline$clone()

Method addPoints()

Add points to the outline register of points

Usage

PathOutline$addPoints(P, fid)

Arguments

Returns

The ID of each added point in the register. If points already exist a point will not be created in the register, but an ID will be returned

Method nextPoint()

Get next point in path for

Usage

PathOutline$nextPoint(pids)

Arguments

Method insertPoint()

Insert point at a fractional distance between points

Usage

PathOutline$insertPoint(i0, i1, f)

Arguments

Method stitchSubpaths()

Stitch subpaths

Usage

PathOutline$stitchSubpaths(VF0, VF1, VB0, VB1, epsilon)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

PathOutline$clone(deep = FALSE)

Arguments

Subclass of FeatureSet to represent points

Description

A PointSet contains information about points located on Outlines. Each PointSet contains a list of matrices, each of which has columns labelled X and Y describing the cartesian coordinates (in the unscaled coordinate frame) of points on the Outline.

Super classes

retistruct::FeatureSetCommon -> retistruct::FeatureSet -> PointSet

Methods

Public methods

• PointSet$new()

• PointSet$reconstruct()

• PointSet$clone()

Method new()

Constructor

Usage

PointSet$new(data = NULL, cols = NULL)

Arguments

Method reconstruct()

Map the PointSet to a ReconstructedOutline

Usage

PointSet$reconstruct(ro)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

PointSet$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Convert an R6 object into a list, ignoring functions and environments

Description

Convert an R6 object into a list, ignoring functions and environments

Usage

R6_to_list(r, path = "", envs = list())

Arguments

Value

List with structure mirroring the R6 object.

Author(s)

David Sterratt

Restore points to spherical manifold

Description

Restore points to spherical manifold after an update of the Lagrange integration rule

Usage

Rcart(P, R, Rset, i0, phi0, lambda0)

Arguments

Value

Points projected back onto sphere

Author(s)

David Sterratt

Class containing functions and data to map CountSets to ReconstructedOutlines

Description

A ReconstructedCountSet contains information about features located on ReconstructedOutlines. Each ReconstructedCountSet contains a list of matrices, each of which has columns labelled phi (latitude) and lambda (longitude) describing the spherical coordinates of points on the ReconstructedOutline, and a column C representing the counts at these points.

Super classes

retistruct::FeatureSetCommon -> retistruct::ReconstructedFeatureSet -> ReconstructedCountSet

Public fields

Methods

Public methods

• ReconstructedCountSet$new()

• ReconstructedCountSet$getKR()

• ReconstructedCountSet$clone()

Method new()

Constructor

Usage

ReconstructedCountSet$new(fs = NULL, ro = NULL)

Arguments

Method getKR()

Get kernel regression estimate of grouped data points

Usage

ReconstructedCountSet$getKR()

Returns

Kernel regression computed using compute.kernel.estimate

Method clone()

The objects of this class are cloneable with this method.

Usage

ReconstructedCountSet$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functions and data to map FeatureSets to ReconstructedOutlines

Description

A ReconstructedFeatureSet contains information about features located on ReconstructedOutlines. Each ReconstructedFeatureSet contains a list of matrices, each of which has columns labelled phi (latitude) and lambda (longitude) describing the spherical coordinates of points on the ReconstructedOutline. Derived classes, e.g. a ReconstructedCountSet, may have extra columns. Each matrix in the list has an associated label and colour, which is used by plotting functions.

Super class

retistruct::FeatureSetCommon -> ReconstructedFeatureSet

Methods

Public methods

• ReconstructedFeatureSet$new()

• ReconstructedFeatureSet$clone()

Method new()

Constructor

Usage

ReconstructedFeatureSet$new(fs = NULL, ro = NULL)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

ReconstructedFeatureSet$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functions and data to map LandmarkSets to ReconstructedOutlines

Description

A ReconstructedLandmarkSet contains information about features located on ReconstructedOutlines. Each ReconstructedLandmarkSet contains a list of matrices, each of which has columns labelled phi (latitude) and lambda (longitude) describing the spherical coordinates of points on the ReconstructedOutline.

Super classes

retistruct::FeatureSetCommon -> retistruct::ReconstructedFeatureSet -> ReconstructedLandmarkSet

Methods

Public methods

• ReconstructedLandmarkSet$clone()

Method clone()

The objects of this class are cloneable with this method.

Usage

ReconstructedLandmarkSet$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functions to reconstruct StitchedOutlines and store the associated data

Description

The function reconstruct reconstructs outline into spherical surface Reconstruct outline into spherical surface.

Super class

retistruct::OutlineCommon -> ReconstructedOutline

Public fields

Methods

Public methods

• ReconstructedOutline$loadOutline()

• ReconstructedOutline$reconstruct()

• ReconstructedOutline$mergePointsEdges()

• ReconstructedOutline$projectToSphere()

• ReconstructedOutline$getStrains()

• ReconstructedOutline$optimiseMapping()

• ReconstructedOutline$optimiseMappingCart()

• ReconstructedOutline$transformImage()

• ReconstructedOutline$getIms()

• ReconstructedOutline$getTearCoords()

• ReconstructedOutline$getFullCutCoords()

• ReconstructedOutline$getNonRimBoundaryCoords()

• ReconstructedOutline$getFeatureSet()

• ReconstructedOutline$reconstructFeatureSets()

• ReconstructedOutline$getPoints()

• ReconstructedOutline$mapFlatToSpherical()

• ReconstructedOutline$titrate()

• ReconstructedOutline$clone()

Method loadOutline()

Load AnnotatedOutline into ReconstructedOutline object

Usage

ReconstructedOutline$loadOutline(
  ol,
  n = 500,
  alpha = 8,
  x0 = 0.5,
  plot.3d = FALSE,
  dev.flat = NA,
  dev.polar = NA,
  report = retistruct::report,
  debug = FALSE
)

Arguments

Method reconstruct()

Reconstruct Reconstruction proceeds in a number of stages:

1. The flat object is triangulated with at least n triangles. This can introduce new vertices in the rim.

2. The triangulated object is stitched.

3. The stitched object is triangulated again, but this time it is not permitted to add extra vertices to the rim.

4. The corresponding points determined by the stitching process are merged to form a new set of merged points and a new triangulation.

5. The merged points are projected roughly to a sphere.

6. The locations of the points on the sphere are moved so as to minimise the energy function.

Usage

ReconstructedOutline$reconstruct(
  plot.3d = FALSE,
  dev.flat = NA,
  dev.polar = NA,
  shinyOutput = NA,
  report = getOption("retistruct.report")
)

Arguments

Method mergePointsEdges()

Merge stitched points and edges. Create merged and transformed versions (all suffixed with t) of a number of existing variables, as well as a matrix Bt, which maps a binary vector representation of edge indices onto a binary vector representation of the indices of the points linked by the edge. Sets following fields

Pt

Transformed point locations

Trt

Transformed triangulation

Ct

Transformed connection set

Cut

Transformed symmetric connection set

Bt

Transformed binary vector representation of edge indices onto a binary vector representation of the indices of the points linked by the edge

Lt

Transformed edge lengths

ht

Transformed correspondences

u

Indices of unique points in untransformed space

U

Transformed indices of unique points in untransformed space

Rset

The set of points on the rim (which has been reordered)

Rsett

Transformed set of points on rim

i0t

Transformed index of the landmark

H

mapping from edges onto corresponding edges

Ht

Transformed mapping from edges onto corresponding edges

Usage

ReconstructedOutline$mergePointsEdges()

Method projectToSphere()

Project mesh points in the flat outline onto a sphere This takes the mesh points from the flat outline and maps them to the curtailed sphere. It uses the area of the flat outline and phi0 to determine the radius R of the sphere. It tries to get a good first approximation by using the function stretchMesh. The following fields are set:

phi

Latitude of mesh points.

lmabda

Longitude of mesh points.

R

Radius of sphere.

Usage

ReconstructedOutline$projectToSphere()

Method getStrains()

Return strains edges are under in spherical retina Set information about how edges on the sphere have been deformed from their flat state.

Usage

ReconstructedOutline$getStrains()

Returns

A list containing two data frames flat and spherical. Each data frame contains for each edge in the flat or spherical meshes:

L

Length of the edge in the flat outline

l

Length of the corresponding edge on the sphere

strain

The strain of each connection

logstrain

The logarithmic strain of each connection

Method optimiseMapping()

Optimise the mapping from the flat outline to the sphere

Usage

ReconstructedOutline$optimiseMapping(
  alpha = 4,
  x0 = 0.5,
  nu = 1,
  optim.method = "BFGS",
  plot.3d = FALSE,
  dev.flat = NA,
  dev.polar = NA,
  shinyOutput = NULL,
  control = list()
)

Arguments

Method optimiseMappingCart()

Optimise the mapping from the flat outline to the sphere

Usage

ReconstructedOutline$optimiseMappingCart(
  alpha = 4,
  x0 = 0.5,
  nu = 1,
  method = "BFGS",
  plot.3d = FALSE,
  dev.flat = NA,
  dev.polar = NA,
  shinyOutput = NULL,
  ...
)

Arguments

Method transformImage()

Transform an image into the reconstructed space Transform an image into the reconstructed space. The four corner coordinates of each pixel are transformed into spherical coordinates and a mask matrix with the same dimensions as im is created. This has TRUE for pixels that should be displayed and FALSE for ones that should not. Sets the field

ims

Coordinates of corners of pixels in spherical coordinates

Usage

ReconstructedOutline$transformImage()

Method getIms()

Get coordinates of corners of pixels of image in spherical coordinates

Usage

ReconstructedOutline$getIms()

Returns

Coordinates of corners of pixels in spherical coordinates

Method getTearCoords()

Get locations of tears in spherical coordinates

Usage

ReconstructedOutline$getTearCoords()

Returns

List containing locations of tears in spherical coordinates

Method getFullCutCoords()

Get locations of fullcuts in spherical coordinates

Usage

ReconstructedOutline$getFullCutCoords()

Returns

List containing locations of fullcuts in spherical coordinates

Method getNonRimBoundaryCoords()

Get location of non-rim boundaries in spherical coordinates

Usage

ReconstructedOutline$getNonRimBoundaryCoords()

Returns

List containing locations of non-rim boundaries in spherical coordinates

Method getFeatureSet()

Get ReconstructedFeatureSet

Usage

ReconstructedOutline$getFeatureSet(type)

Arguments

Method reconstructFeatureSets()

Reconstruct any attached feature sets.

Usage

ReconstructedOutline$reconstructFeatureSets()

Method getPoints()

Get mesh points in spherical coordinates

Usage

ReconstructedOutline$getPoints()

Returns

Matrix with columns phi (latitude) and lambda (longitude)

Method mapFlatToSpherical()

Return location of point on sphere corresponding to point on the flat outline

Usage

ReconstructedOutline$mapFlatToSpherical(P)

Arguments

Method titrate()

Try a range of values of phi0s in the reconstruction, recording the energy of the mapping in each case.

Usage

ReconstructedOutline$titrate(
  alpha = 8,
  x0 = 0.5,
  byd = 1,
  len.up = 5,
  len.down = 20
)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

ReconstructedOutline$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functions and data to map PointSets to ReconstructedOutlines

Description

A ReconstructedPointSet contains information about features located on ReconstructedOutlines. Each ReconstructedPointSet contains a list of matrices, each of which has columns labelled phi (latitude) and lambda (longitude) describing the spherical coordinates of points on the ReconstructedOutline.

Super classes

retistruct::FeatureSetCommon -> retistruct::ReconstructedFeatureSet -> ReconstructedPointSet

Public fields

Methods

Public methods

• ReconstructedPointSet$getMean()

• ReconstructedPointSet$getHullarea()

• ReconstructedPointSet$getKDE()

• ReconstructedPointSet$clone()

Method getMean()

Get Karcher mean of datapoints in spherical coordinates

Usage

ReconstructedPointSet$getMean()

Returns

Karcher mean of datapoints in spherical coordinates

Method getHullarea()

Get area of convex hull around data points on sphere

Usage

ReconstructedPointSet$getHullarea()

Returns

Area in degrees squared

Method getKDE()

Get kernel density estimate of data points

Usage

ReconstructedPointSet$getKDE()

Returns

See compute.kernel.estimate

Method clone()

The objects of this class are cloneable with this method.

Usage

ReconstructedPointSet$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functions and data relating to retinal outlines

Description

In addition to fields inherited from StitchedOutline, a RetinalOutline contains a dataset field, describing the system path to dataset directory and metadata specific to retinae and some formats of retinae

An retinalOutline object. This contains the following fields:

Super classes

retistruct::OutlineCommon -> retistruct::Outline -> retistruct::PathOutline -> retistruct::AnnotatedOutline -> retistruct::TriangulatedOutline -> retistruct::StitchedOutline -> RetinalOutline

Public fields

Methods

Public methods

• RetinalOutline$new()

• RetinalOutline$clone()

Method new()

Constructor

Usage

RetinalOutline$new(..., dataset = NULL)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

RetinalOutline$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

A version of ReconstructedOutline that is specific to retinal datasets

Description

A RetinalReconstructedOutline overrides methods of ReconstructedOutline so that they return data point and landmark coordinates that have been transformed according to the values of DVflip and side. When reconstructing, it also computes the “Optic disc displacement”, i.e. the number of degrees subtended between the optic disc and the pole.

Super classes

retistruct::OutlineCommon -> retistruct::ReconstructedOutline -> RetinalReconstructedOutline

Public fields

Methods

Public methods

• RetinalReconstructedOutline$getIms()

• RetinalReconstructedOutline$getTearCoords()

• RetinalReconstructedOutline$reconstruct()

• RetinalReconstructedOutline$getFeatureSet()

• RetinalReconstructedOutline$clone()

Method getIms()

Get coordinates of corners of pixels of image in spherical coordinates, transformed according to the value of DVflip

Usage

RetinalReconstructedOutline$getIms()

Returns

Coordinates of corners of pixels in spherical coordinates

Method getTearCoords()

Get location of tear coordinates in spherical coordinates, transformed according to the value of DVflip

Usage

RetinalReconstructedOutline$getTearCoords()

Returns

Location of tear coordinates in spherical coordinates

Method reconstruct()

Usage

RetinalReconstructedOutline$reconstruct(...)

Arguments

Method getFeatureSet()

Get ReconstructedFeatureSet, transformed according to the value of DVflip

Usage

RetinalReconstructedOutline$getFeatureSet(type)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

RetinalReconstructedOutline$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functions and data relating to Stitching outlines

Description

A StitchedOutline contains a function to stitch the tears and fullcuts, setting the correspondences hf, hb and h

Super classes

retistruct::OutlineCommon -> retistruct::Outline -> retistruct::PathOutline -> retistruct::AnnotatedOutline -> retistruct::TriangulatedOutline -> StitchedOutline

Public fields

Methods

Public methods

• StitchedOutline$new()

• StitchedOutline$stitchTears()

• StitchedOutline$stitchFullCuts()

• StitchedOutline$isStitched()

• StitchedOutline$getBoundarySets()

• StitchedOutline$clone()

Method new()

Constructor

Usage

StitchedOutline$new(...)

Arguments

Method stitchTears()

Stitch together the incisions and tears by inserting new points in the tears and creating correspondences between new points.

Usage

StitchedOutline$stitchTears()

Method stitchFullCuts()

Stitch together the fullcuts by inserting new points in the tears and creating correspondences between new points.

Usage

StitchedOutline$stitchFullCuts()

Method isStitched()

Test if the outline has been stitched

Usage

StitchedOutline$isStitched()

Returns

Boolean, indicating if the outline has been stitched or not

Method getBoundarySets()

Get point IDs of points on boundaries

Usage

StitchedOutline$getBoundarySets()

Returns

List of Point IDs of points on the boundaries. If the outline has been stitched, the point IDs in each element of the list will be ordered in the direction of the forward pointer, and the boundary that is longest will be named as Rim. If the outline has not been stitched, the list will have one element named Rim.

Method clone()

The objects of this class are cloneable with this method.

Usage

StitchedOutline$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class to triangulate Fragments

Description

A TriangulatedFragment contains a function to create a triangulated mesh over an fragment, and fields to hold the mesh information.

Super class

retistruct::Fragment -> TriangulatedFragment

Public fields

Methods

Public methods

• TriangulatedFragment$new()

• TriangulatedFragment$clone()

Method new()

Constructor

Usage

TriangulatedFragment$new(
  fragment,
  n = 200,
  suppress.external.steiner = FALSE,
  report = message
)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

TriangulatedFragment$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Class containing functions and data relating to Triangulation

Description

A TriangulatedOutline contains a function to create a triangulated mesh over an outline, and fields to hold the mesh information. Note that areas and lengths are all scaled using the value of the scale field.

Super classes

retistruct::OutlineCommon -> retistruct::Outline -> retistruct::PathOutline -> retistruct::AnnotatedOutline -> TriangulatedOutline

Public fields

Methods

Public methods

• TriangulatedOutline$triangulate()

• TriangulatedOutline$mapTriangulatedFragment()

• TriangulatedOutline$clone()

Method triangulate()

Triangulate (mesh) outline

Usage

TriangulatedOutline$triangulate(n = 200, suppress.external.steiner = FALSE)

Arguments

Method mapTriangulatedFragment()

Map the point IDs of a TriangulatedFragment on the point IDs of this Outline

Usage

TriangulatedOutline$mapTriangulatedFragment(fragment, pids)

Arguments

Method clone()

The objects of this class are cloneable with this method.

Usage

TriangulatedOutline$clone(deep = FALSE)

Arguments

Author(s)

David Sterratt

Examples

P <- rbind(c(1,1),   c(2,1),  c(2,-1),
           c(1,-1),  c(1,-2), c(-1,-2),
           c(-1,-1), c(-2,-1),c(-2,1),
           c(-1,1),  c(-1,2), c(1,2))
o <- TriangulatedOutline$new(P)
o$addTear(c(3, 4, 5))
o$addTear(c(6, 7, 8))
o$addTear(c(9, 10, 11))
o$addTear(c(12, 1, 2))
flatplot(o)

P <- list(rbind(c(1,1), c(2,1), c(2.5,2), c(3,1), c(4,1), c(1,4)),
              rbind(c(-1,1), c(-1,4), c(-2,3), c(-2,2), c(-3,2), c(-4,1)),
              rbind(c(-4,-1), c(-1,-1), c(-1,-4)),
              rbind(c(1,-1), c(2,-1), c(2.5,-2), c(3,-1), c(4,-1), c(1,-4)))
o <- TriangulatedOutline$new(P)
##' o$addTear(c(2, 3, 4))
o$addTear(c(17, 18, 19))
o$addTear(c(9, 10, 11))
o$addFullCut(c(1, 5, 16, 20))
flatplot(o)

Convert azimuth-elevation coordinates to spherical coordinates

Description

Convert azimuth-elevation coordinates to spherical coordinates

Usage

azel.to.sphere.colatitude(r, r0)

Arguments

Value

2-column matrix of spherical coordinates of points with column names psi (colatitude) and lambda (longitude).

Author(s)

David Sterratt

Examples

r0 <- cbind(alpha=0, theta=0)
r <- rbind(r0, r0+c(1,0), r0-c(1,0), r0+c(0,1), r0-c(0,1))
azel.to.sphere.colatitude(r, r0)

Azimuthal conformal or stereographic or Wulff projection

Description

Azimuthal conformal or stereographic or Wulff projection

Usage

azimuthal.conformal(r, ...)

Arguments

Value

2-column Matrix of Cartesian coordinates of points on polar projection. Column names should be x and y.

Note

This is a special case with the point centred on the projection being the South Pole. The MathWorld equations are for the more general case.

Author(s)

David Sterratt

References

https://en.wikipedia.org/wiki/Map_projection, http://mathworld.wolfram.com/StereographicProjection.html Fisher, N. I., Lewis, T., and Embleton, B. J. J. (1987). Statistical analysis of spherical data. Cambridge University Press, Cambridge, UK.

Lambert azimuthal equal area projection

Description

Lambert azimuthal equal area projection

Usage

azimuthal.equalarea(r, ...)

Arguments

Value

2-column Matrix of Cartesian coordinates of points on polar projection. Column names should be x and y.

Note

This is a special case with the point centred on the projection being the South Pole. The MathWorld equations are for the more general case.

Author(s)

David Sterratt

References

https://en.wikipedia.org/wiki/Map_projection, http://mathworld.wolfram.com/LambertAzimuthalEqual-AreaProjection.html Fisher, N. I., Lewis, T., and Embleton, B. J. J. (1987). Statistical analysis of spherical data. Cambridge University Press, Cambridge, UK.

Azimuthal equidistant projection

Description

Azimuthal equidistant projection

Usage

azimuthal.equidistant(r, ...)

Arguments

Value

2-column Matrix of Cartesian coordinates of points on polar projection. Column names should be x and y.

Note

This is a special case with the point centred on the projection being the South Pole. The MathWorld equations are for the more general case.

Author(s)

David Sterratt

References

https://en.wikipedia.org/wiki/Map_projection, http://mathworld.wolfram.com/AzimuthalEquidistantProjection.html

Convert barycentric coordinates of points in mesh on sphere to cartesian coordinates

Description

Given a triangular mesh on a sphere described by mesh locations (phi, lambda), a radius R and a triangulation Trt, determine the Cartesian coordinates of points cb given in barycentric coordinates with respect to the mesh.

Usage

bary.to.sphere.cart(phi, lambda, R, Trt, cb)

Arguments

Value

An N-by-3 matrix of the Cartesian coordinates of the points

Author(s)

David Sterratt

Central angle between two points on a sphere

Description

On a sphere the central angle between two points is defined as the angle whose vertex is the centre of the sphere and that subtends the arc formed by the great circle between the points. This function computes the central angle for two points (\phi_1, \lambda_1) and (\phi_2,\lambda_2).

Usage

central.angle(phi1, lambda1, phi2, lambda2)

Arguments

Value

Central angle

Author(s)

David Sterratt

Source

Wikipedia https://en.wikipedia.org/wiki/Central_angle

Check the whether directory contains valid data

Description

Check the whether directory contains valid data

Usage

checkDatadir(dir = NULL)

Arguments

Value

TRUE if dir contains valid data; FALSE otherwise.

Author(s)

David Sterratt

Return points on the unit circle

Description

Return points on the unit circle in an anti-clockwise direction. If L is not specified n points are returned. If L is specified, the same number of points are returned as there are elements in L, the interval between successive points being proportional to L.

Usage

circle(n = 12, L = NULL)

Arguments

Value

The cartesian coordinates of the points

Author(s)

David Sterratt

Find the intersection of a plane with edges of triangles on a sphere

Description

Find the intersections of the plane defined by the normal n and the distance d expressed as a fractional distance along the side of each triangle.

Usage

compute.intersections.sphere(phi, lambda, T, n, d)

Arguments

Value

Matrix with same dimensions as T. Each row gives the intersection of the plane with the corresponding triangle in T. Column 1 gives the fractional distance from vertex 2 to vertex 3. Column 2 gives the fractional distance from vertex 3 to vertex 1. Column 2 gives the fractional distance from vertex 1 to vertex 2. A value of NaN indicates that the corresponding edge lies in the plane. A value of Inf indicates that the edge lies parallel to the plane but outside it.

Author(s)

David Sterratt

Kernel estimate over grid

Description

Compute a kernel estimate over a grid and do a contour analysis of this estimate. The contour heights the determined by finding heights that exclude a certain fraction of the probability. For example, the 95 and it should enclose about 5 are specified by the contour.levels option; by default they are c(5, 25, 50, 75, 95).

Usage

compute.kernel.estimate(Dss, phi0, fhat, compute.conc)

Arguments

Value

A list containing

Author(s)

David Sterratt

Create grid on projection of hemisphere onto plane

Description

Create grid on projection of hemisphere onto plane

Usage

create.polar.cart.grid(pa, res, phi0)

Arguments

Value

List containing:

Author(s)

David Sterratt

Read a retinal dataset in CSV format

Description

Read a retinal dataset in CSV format. Each dataset is a folder containing a file called outline.csv that specifies the outline in X-Y coordinates. It may also contain a file datapoints.csv, containing the locations of data points and a file datacounts.csv, containing the locations of data counts; see read.datapoints and read.datacounts for the formats of these files. The folder may also contain a file od.csv specifying the coordinates of the optic disc.

Usage

csv.read.dataset(dataset, report = message)

Arguments

Value

A RetinalOutline object

Author(s)

David Sterratt

The deformation energy gradient function

Description

The function that computes the gradient of the energy (or error) of the deformation of the mesh from the flat outline to the sphere. This depends on the locations of the points given in spherical coordinates. The function is designed to take these as a vector that is received from the optim function.

Usage

dE(
  p,
  Cu,
  C,
  L,
  B,
  Tr,
  A,
  R,
  Rset,
  i0,
  phi0,
  lambda0,
  Nphi,
  N,
  alpha = 1,
  x0,
  nu = 1,
  verbose = FALSE
)

Arguments

Value

A vector representing the derivative of the energy of this particular configuration with respect to the parameter vector

Author(s)

David Sterratt

Draw the "flat" outline in 3D with depth information

Description

Draw the "flat" outline in 3D with depth information

Usage

depthplot3D(r, ...)

Arguments

Author(s)

David Sterratt

File system directories used by shinyFiles

Description

File system directories used by shinyFiles

Usage

directories()

Piecewise smooth function used in area penalty

Description

Piecewise, smooth function that increases linearly with negative arguments.

f(x) = \left\{ \begin{array}{ll} -(x - x_0/2) & x < 0 \\ \frac{1}{2x_0}(x - x_0)^2 & 0 < x <x_0 \\ 0 & x \ge x_0 \end{array} \right.

Usage

f(x, x0)

Arguments

Value

The value of the function.

Author(s)

David Sterratt

The FIRE algorithm

Description

This is an implementation of the FIRE algorithm for structural relaxation put forward by Bitzek et al. (2006)

Usage

fire(
  r,
  force,
  restraint,
  m = 1,
  dt = 0.1,
  maxmove = 100,
  dtmax = 1,
  Nmin = 5,
  finc = 1.1,
  fdec = 0.5,
  astart = 0.1,
  fa = 0.99,
  a = 0.1,
  nstep = 100,
  tol = 1e-05,
  verbose = FALSE,
  report = message
)

Arguments

Value

List containing x, the positions of the points, conv, which is 0 if convergence as occurred and 1 otherwise, and frms, the root mean square of the forces on the particles.

Author(s)

David Sterratt

References

Bitzek, E., Koskinen, P., Gähler, F., Moseler, M., and Gumbsch, P. (2006). Structural relaxation made simple. Phys. Rev. Lett., 97:170201.

Plot "flat" (unreconstructed) representation of outline

Description

Plot "flat" (unreconstructed) representation of outline

Usage

flatplot(x, ...)

Arguments

Author(s)

David Sterratt

Flat plot of AnnotatedOutline

Description

Plot flat AnnotatedOutline. The user markup is displayed by default.

Usage

## S3 method for class 'AnnotatedOutline'
flatplot(x, axt = "n", xlim = NULL, ylim = NULL, markup = TRUE, ...)

Arguments

Author(s)

David Sterratt

Flat plot of outline

Description

Plot flat Outline.

Usage

## S3 method for class 'Outline'
flatplot(
  x,
  axt = "n",
  xlim = NULL,
  ylim = NULL,
  add = FALSE,
  image = TRUE,
  scalebar = 1,
  rimset = FALSE,
  pids = FALSE,
  pid.joggle = 0,
  lwd.outline = 1,
  ...
)

Arguments

Author(s)

David Sterratt

Flat plot of reconstructed outline

Description

Plot ReconstructedOutline object. This adds a mesh of gridlines from the spherical retina (described by points phi, lambda and triangulation Trt and cut-off point phi0) onto a flattened retina (described by points P and triangulation T).

Usage

## S3 method for class 'ReconstructedOutline'
flatplot(
  x,
  axt = "n",
  xlim = NULL,
  ylim = NULL,
  grid = TRUE,
  strain = FALSE,
  ...
)

Arguments

Author(s)

David Sterratt

Flat plot of AnnotatedOutline

Description

Plot flat StitchedOutline. If the optional argument stitch is TRUE the user markup is displayed.

Usage

## S3 method for class 'StitchedOutline'
flatplot(x, axt = "n", xlim = NULL, ylim = NULL, stitch = TRUE, lwd = 1, ...)

Arguments

Author(s)

David Sterratt

Plot flat TriangulatedOutline.

Description

Plot flat TriangulatedOutline.

Usage

## S3 method for class 'TriangulatedOutline'
flatplot(x, axt = "n", xlim = NULL, ylim = NULL, mesh = TRUE, ...)

Arguments

Author(s)

David Sterratt

Determine indices of triangles that are flipped

Description

In the projection of points onto the sphere, some triangles maybe flipped, i.e. in the wrong orientation. This functions determines which triangles are flipped by computing the vector pointing to the centre of each triangle and comparing this direction to vector product of two sides of the triangle.

Usage

flipped.triangles(Ps, Trt, R = 1)

Arguments

Value

List containing:

Author(s)

David Sterratt

Determine indices of triangles that are flipped

Description

In the projection of points onto the sphere, some triangles maybe flipped, i.e. in the wrong orientation. This function determines which triangles are flipped by computing the vector pointing to the centre of each triangle and comparing this direction to vector product of two sides of the triangle.

Usage

flipped.triangles.cart(P, Trt, R)

Arguments

Value

List containing:

Author(s)

David Sterratt

Piecewise smooth function used in area penalty

Description

Derivative of f

Usage

fp(x, x0)

Arguments

Value

The value of the function.

Author(s)

David Sterratt

The identity transformation

Description

The identity transformation

Usage

identity.transform(r, ...)

Arguments

Value

Identical matrix

Author(s)

David Sterratt

Read one of the Thompson lab's retinal datasets

Description

Read one of the Thompson lab's retinal datasets. Each dataset is a folder containing a SYS file in SYSTAT format and a MAP file in text format. The SYS file specifies the locations of the data points and the MAP file specifies the outline.

Usage

idt.read.dataset(dataset, report = message, d.close = 0.25)

Arguments

Details

The function returns the outline of the retina. In order to do so, it has to join up the segments of the MAP file. The tracings are not always precise; sometimes there are gaps between points that are actually the same point. The parameter d.close specifies how close points must be to count as the same point.

Value

Author(s)

David Sterratt

Read a retinal dataset in IJROI format

Description

Read a retinal dataset in IJROI format. Each dataset is a folder containing a file called outline.roi that specifies the outline in X-Y coordinates. It may also contain a file datapoints.csv, containing the locations of data points; see read.datapoints for the format of this file. The folder may also contain a file od.roi specifying the coordinates of the optic disc.

Usage

ijroi.read.dataset(dataset, report = report)

Arguments

Value

A RetinalOutline object

Author(s)

David Sterratt

Read a retinal dataset in IJROI format

Description

Read a retinal dataset in IJROI format. Each dataset is a folder containing a file called outline.roi that specifies the outline in X-Y coordinates. It may also contain a file datapoints.csv, containing the locations of data points; see read.datapoints for the format of this file. The folder may also contain a file od.roi specifying the coordinates of the optic disc.

Usage

ijroimulti.read.dataset(dataset)

Arguments

Value

A RetinalOutline object

Author(s)

David Sterratt

Interpolate values in image

Description

Interpolate values in image

Usage

interpolate.image(im, P, invert.y = FALSE, wmin = 10, wmax = 100)

Arguments

Value

Vector of N interpolated values

Author(s)

David Sterratt

Invert sphere about its centre

Description

Invert sphere about its centre

Usage

invert.sphere(r, ...)

Arguments

Value

Matrix in same format, but with pi added to lambda and phi negated.

Author(s)

David Sterratt

Invert sphere to hemisphere

Description

Invert image of a partial sphere and scale the longitude so that points at latitude phi0 is projected onto a longitude of 0 degrees (the equator).

Usage

invert.sphere.to.hemisphere(r, phi0, ...)

Arguments

Value

Matrix in same format, but with pi added to lambda and phi negated and scaled so that the longitude phi0 is projected to 0 degrees (the equator)

Author(s)

David Sterratt

Karcher mean on the sphere

Description

The Karcher mean of a set of points on a manifold is defined as the point whose sum of squared Riemann distances to the points is minimal. On a sphere using spherical coordinates this distance can be computed using the formula for central angle.

Usage

karcher.mean.sphere(x, na.rm = FALSE, var = FALSE)

Arguments

Value

Vector of means with components named phi and lambda. If var is TRUE, a list containing mean and variance in elements mean and var.

Author(s)

David Sterratt

References

Heo, G. and Small, C. G. (2006). Form representations and means for landmarks: A survey and comparative study. Computer Vision and Image Understanding, 102:188-203.

See Also

central.angle

Estimate of the log likelihood of the points mu given a particular value of the concentration kappa

Description

Estimate of the log likelihood of the points mu given a particular value of the concentration kappa

Usage

kde.L(mu, kappa)

Arguments

Value

Log likelihood of data

Author(s)

David Sterratt

Find the optimal concentration for a set of data

Description

Find the optimal concentration for a set of data

Usage

kde.compute.concentration(mu)

Arguments

Value

The optimal concentration

Author(s)

David Sterratt

Kernel density estimate on sphere using Fisherian density with polar coordinates

Description

Kernel density estimate on sphere using Fisherian density with polar coordinates

Usage

kde.fhat(r, mu, kappa)

Arguments

Value

Vector of density estimates

Author(s)

David Sterratt

Kernel density estimate on sphere using Fisherian density with Cartesian coordinates

Description

Kernel density estimate on sphere using Fisherian density with Cartesian coordinates

Usage

kde.fhat.cart(r, mu, kappa)

Arguments

Value

Vector of density estimates

Author(s)

David Sterratt

Find the optimal concentration for a set of data

Description

Find the optimal concentration for a set of data

Usage

kr.compute.concentration(mu, y)

Arguments

Value

The optimal concentration

Author(s)

David Sterratt

Cross validation estimate of the least squares error of the points mu given a particular value of the concentration kappa

Description

Cross validation estimate of the least squares error of the points mu given a particular value of the concentration kappa

Usage

kr.sscv(mu, y, kappa)

Arguments

Value

Least squares error

Author(s)

David Sterratt

Kernel regression on sphere using Fisherian density with polar coordinates

Description

Kernel regression on sphere using Fisherian density with polar coordinates

Usage

kr.yhat(r, mu, y, kappa)

Arguments

Value

Estimates of dependent variables at locations r

Author(s)

David Sterratt

Kernel regression on sphere using Fisherian density with Cartesian coordinates

Description

Kernel regression on sphere using Fisherian density with Cartesian coordinates

Usage

kr.yhat.cart(r, mu, y, kappa)

Arguments

Value

Estimates of dependent variables at locations r

Author(s)

David Sterratt

Determine intersection between two lines

Description

Determine the intersection of two lines L1 and L2 in two dimensions, using the formula described by Weisstein.

Usage

line.line.intersection(P1, P2, P3, P4, interior.only = FALSE)

Arguments

Value

Vector containing x,y coordinates of intersection of L1 and L2. If L1 and L2 are parallel, this is infinite-valued. If interior.only is TRUE, then when the intersection does not occur between P1 and P2 and P3 and P4, a vector containing NAs is returned.

Author(s)

David Sterratt

Source

Weisstein, Eric W. "Line-Line Intersection." From MathWorld–A Wolfram Web Resource. http://mathworld.wolfram.com/Line-LineIntersection.html

Examples

## Intersection of two intersecting lines
line.line.intersection(c(0, 0), c(1, 1), c(0, 1), c(1, 0))

## Two lines that don't intersect
line.line.intersection(c(0, 0), c(0, 1), c(1, 0), c(1, 1))

List datasets underneath a directory

Description

List valid datasets underneath a directory. This reports all directories that appear to be valid.

Usage

list.datasets(path = ".", verbose = FALSE)

Arguments

Value

A vector of directories containing datasets

Author(s)

David Sterratt

Convert an list created by R6_to_list() into an R6 object.

Description

Convert an list created by R6_to_list() into an R6 object.

Usage

list_to_R6(l)

Arguments

Value

R6 object or list list

Author(s)

David Sterratt

Plot the fractional change in length of mesh edges

Description

Plot the fractional change in length of mesh edges. The length of each edge in the mesh in the reconstructed object is plotted against each edge in the spherical object. The points are colour-coded according to the amount of log strain each edge is under.

Usage

lvsLplot(r, ...)

Arguments

Author(s)

David Sterratt

Morph a flat dataset to a parabola for testing purposes

Description

Morph a flat dataset to a parabola for testing purposes

Usage

morph.dataset.to.parabola(
  orig.dataset = file.path(system.file(package = "retistruct"), "extdata", "smi32-csv"),
  morphed.dataset = NA,
  f = 100
)

Arguments

Value

Path to morphed.dataset

Author(s)

David Sterratt

Return a new version of the list in which any unnamed elements have been given standardised names

Description

Return a new version of the list in which any unnamed elements have been given standardised names

Usage

name.list(l)

Arguments

Value

The list with standardised names

Author(s)

David Sterratt

Bring angle into range

Description

Bring angle into range

Usage

normalise.angle(theta)

Arguments

Value

Normalised angle

Author(s)

David Sterratt

Orthographic projection

Description

Orthographic projection

Usage

orthographic(r, proj.centre = cbind(phi = 0, lambda = 0), ...)

Arguments

Value

Two-column matrix with columns labelled x and y of locations of projection of coordinates on plane

Author(s)

David Sterratt

References

https://en.wikipedia.org/wiki/Map_projection, http://mathworld.wolfram.com/OrthographicProjection.html

Ancillary function to place labels

Description

Ancillary function to place labels

Usage

panlabel(panlabel, line = -0.7)

Arguments

Author(s)

David Sterratt

Arc length of a parabola y=x^2/4f

Description

Arc length of a parabola y=x^2/4f

Usage

parabola.arclength(x1, x2, f)

Arguments

Value

length of parabola arc

Author(s)

David Sterratt

Inverse arc length of a parabola y=x^2/4f

Description

Inverse arc length of a parabola y=x^2/4f

Usage

parabola.invarclength(x1, s, f)

Arguments

Value

x co-ordinate of end of arc

Author(s)

David Sterratt

Parse dependencies

Description

Parse dependencies

Usage

parse.dependencies(deps)

Arguments

Value

Table with package column, relationship column and version number

Author(s)

David Sterratt

Convert polar projection in Cartesian coordinates to spherical coordinates on sphere

Description

This is the inverse of sphere.spherical.to.polar.cart

Usage

polar.cart.to.sphere.spherical(r, pa = FALSE, preserve = "latitude")

Arguments

Value

2-column Matrix of spherical coordinates of points on sphere. Column names are phi and lambda.

Author(s)

David Sterratt

Put text on the polar plot

Description

Place text at bottom right of projection

Usage

polartext(text)

Arguments

Author(s)

David Sterratt

Plot projection of a reconstructed outline

Description

Plot projection of a reconstructed outline

Usage

projection(r, ...)

Arguments

Author(s)

David Sterratt

Projection of a reconstructed outline

Description

Draw a projection of a ReconstructedOutline. This method sets up the grid lines and the angular labels and draws the image.

Usage

## S3 method for class 'ReconstructedOutline'
projection(
  r,
  transform = identity.transform,
  axisdir = cbind(phi = 90, lambda = 0),
  projection = azimuthal.equalarea,
  proj.centre = cbind(phi = 0, lambda = 0),
  lambdalim = c(-180, 180),
  philim = c(-90, 90),
  labels = c(0, 90, 180, 270),
  mesh = FALSE,
  grid = TRUE,
  grid.bg = "transparent",
  grid.int.minor = 15,
  grid.int.major = 45,
  colatitude = TRUE,
  pole = FALSE,
  image = TRUE,
  markup = TRUE,
  add = FALSE,
  max.proj.dim = getOption("max.proj.dim"),
  ...
)

Arguments

Plot projection of reconstructed dataset

Description

Plot projection of reconstructed dataset

Usage

## S3 method for class 'RetinalReconstructedOutline'
projection(
  r,
  transform = identity.transform,
  projection = azimuthal.equalarea,
  axisdir = cbind(phi = 90, lambda = 0),
  proj.centre = cbind(phi = 0, lambda = 0),
  lambdalim = c(-180, 180),
  datapoints = TRUE,
  datapoint.means = TRUE,
  datapoint.contours = FALSE,
  grouped = FALSE,
  grouped.contours = FALSE,
  landmarks = TRUE,
  mesh = FALSE,
  grid = TRUE,
  image = TRUE,
  ids = r$getIDs(),
  ...
)

Arguments

Read data counts in CSV format

Description

Read data counts from a file ‘datacounts.csv’ in the directory dataset. The CSV file should contain two columns for every dataset. Each pair of columns must contain a unique name in the first cell of the first row and a valid colour in the second cell of the first row. In the remaining rows, the X coordinates of data counts should be in the first column and the Y coordinates should be in the second column.

Usage

read.datacounts(dataset)

Arguments

Value

List containing

Author(s)

David Sterratt

Read data points in CSV format

Description

Read data points from a file dataponts.csv in the directory dataset. The CSV should contain two columns for every dataset. Each pair of columns must contain a unique name in the first cell of the first row and a valid colour in the second cell of the first row. In the remaining rows, the X coordinates of data points should be in the first column and the Y coordinates should be in the second column.

Usage

read.datapoints(dataset)

Arguments

Value

List containing

Author(s)

David Sterratt

Remove identical consecutive rows from a matrix

Description

This is similar to unique(), but spares rows which are duplicated, but at different points in the matrix

Usage

remove.identical.consecutive.rows(P)

Arguments

Value

Matrix with identical consecutive rows removed.

Author(s)

David Sterratt

Remove intersections between adjacent segments in a closed path

Description

Suppose segments AB and CD intersect. Point B is replaced by the intersection point, defined B'. Point C is replaced by a point C' on the line B'D. The maximum distance of B'C' is given by the parameter d. If the distance l B'D is less than 2d, the distance B'C' is l/2.

Usage

remove.intersections(P, d = 50)

Arguments

Value

A new closed path without intersections

Author(s)

David Sterratt

Reporting utility function

Description

Calls function specified by option retistruct.report

Usage

report(x, ...)

Arguments

Author(s)

David Sterratt

Start the Retistruct GUI

Description

Start the Retistruct GUI

Usage

retistruct()

Value

Object with getData() method to return reconstructed retina data and environment this which contains variables in object.

Batch operation using the parallel package

Description

This function reconstructs a number of datasets, using the R parallel package to distribute the reconstruction of multiple datasets across CPUs. If datasets is not specified the function recurses through a directory tree starting at tldir, determining whether the directory contains valid raw data and markup, and performing the reconstruction if it does.

Usage

retistruct.batch(
  tldir = ".",
  outputdir = tldir,
  datasets = NULL,
  device = "pdf",
  titrate = FALSE,
  cpu.time.limit = 3600,
  mc.cores = getOption("mc.cores", 2L)
)

Arguments

Author(s)

David Sterratt

Export data from reconstruction data files to MATLAB

Description

Recurse through a directory tree, determining whether the directory contains valid derived data and converting ‘r.rData’ files to files in MATLAB format named ‘r.mat’

Usage

retistruct.batch.export.matlab(tldir = ".")

Arguments

Author(s)

David Sterratt

Plot figures for a batch of reconstructions

Description

Recurse through a directory tree, determining whether the directory contains valid derived data and plotting graphs if it does.

Usage

retistruct.batch.figures(tldir = ".", outputdir = tldir, ...)

Arguments

Author(s)

David Sterratt

Get titrations from a directory of reconstructions

Description

Get titrations from a directory of reconstructions

Usage

retistruct.batch.get.titrations(tldir = ".")

Arguments

Plot titrations

Description

Plot titrations

Usage

retistruct.batch.plot.titrations(tdat)

Arguments

Extract summary data for a batch of reconstructions

Description

Recurse through a directory tree, determining whether the directory contains valid derived data and extracting summary data if it does.

Usage

retistruct.batch.summary(tldir = ".", cache = TRUE)

Arguments

Value

Data frame containing summary data

Author(s)

David Sterratt

Retistruct check markup

Description

Check that markup such as tears and the nasal or dorsal points are present.

Usage

retistruct.check.markup(o)

Arguments

Value

If all markup is present, return TRUE. Otherwise return FALSE.

Author(s)

David Sterratt

Process a dataset with a time limit

Description

This calls retistruct.cli.process with a time limit specified by cpu.time.limit.

Usage

retistruct.cli(
  dataset,
  cpu.time.limit = Inf,
  outputdir = NA,
  device = "pdf",
  ...
)

Arguments

Value

A list comprising

Author(s)

David Sterratt

Print a figure to file

Description

Print a figure to file

Usage

retistruct.cli.figure(
  dataset,
  outputdir,
  device = "pdf",
  width = 6,
  height = 6,
  res = 100
)

Arguments

Author(s)

David Sterratt

Process a dataset, saving results to disk

Description

This function processes a dataset, saving the reconstruction data and MATLAB export data to the dataset directory and printing figures to outputdir.

Usage

retistruct.cli.process(
  dataset,
  outputdir = NA,
  device = "pdf",
  titrate = FALSE,
  matlab = TRUE
)

Arguments