bezier

Db = bezier(D::GMTdataset; t=nothing, np::Int=0, pure=false, firstcurve=true) -> GMTdataset

mat = bezier(p::Matrix{<:Real}; t=nothing, np::Int=0, pure=false, firstcurve=true) -> Matrix{Float64}

mat = bezier(p1::Vector{T}, p2::Vector{T}, p3::Vector{T}, p4::Vector{T}; t=nothing, np::Int=0, pure=false) where {T <: Real}

Create a Bezier curve from a set of control points.

A function for cubic Bezier interpolation for a given set of control points (knots). Each control point can exist in an N-dimensional space but only tested 2D and 3D cases. Data can be geographic or Cartesian. The default for this function is to compute a curve that passes through the control points. This can be changed by setting pure to true and then we instead compute a curve that passes through the first and last.

When more than 4 control points are provided, we make a curve composition by connecting the first curve, made with points 1,2,3,4, second curve, made with points 2,3,4,5 and so on. The way this connection is made is controlled by the firstcurve argument. With more than 4 control points, the pure option is not allowed.

Args

D: A GMTdataset object with a matrix of control points. This matrix may have 2 (2D) or 3 columns (3D). If D represents geographical data, we create a spherical Bezier curve by using an intermediate step of converting the data to ECEF, compute the 3D curve, and then convert it back to geographic coordinates.

p::Matrix{<:Real}: Matrix with at least 4 control points. Same as the D case, except for the possibility of geographical data.

p1, p2, p3, p4: Vectors of control points. They can be 2D or 3D (two or three elements).

Kwargs

t: Vector of values of the t paramter at which bezier curve is evaluated. By default is evaluated at 101 evenly spaced values between 0 and 1, but see np below.
np: Number of values of the t paramter at which bezier curve is evaluated. Default is 101.
pure: By default we compute a curve that passes through the control points. If pure is true, we instead compute a curve that passes through the first and last control points and the other two are used to control the path. This is what pure cubic bezier curves do.
firstcurve: Applies only when the number of control points is > 4 and sets how the different curves are connected. If true, the first curve takes precedence. Using as example the case of two curves given by 5 control points; the final curve is obtained by connecting the first curve (made with the first 4 control points) with the second curve (made with the last 4 control points). But since these two curves shared the second, third and fourth control points, that part is removed from the second curve, leaving only the 4rth to 5th contribution. When firstcurve is false, the opposite happens. The first curve contributes only with the part from first to second control points.

Returns

Db: A GMTdataset object with the coordinates of the Bezier curve, when input is a GMTdataset.
mat::Matrix{Float64}: Matrix of coordinates of the Bezier curve, when input is a matrix.

Example

# A spherical Bezier curve with 4 knots
D = bezier(mat2ds([30. -15; 30 45; -30 45; -30 -15], proj4="geog"));

# A composed spherical Bezier curve with 5 knots and predominance  of second curve
D = bezier(mat2ds([30. -15; 30 45; -30 45; -30 -15; -20 -25], proj4="geog"), firstcurve=false);

Args

Kwargs

Returns

Example

See Also