Getting Started

Documentation

Glish

Learn More

Programming

Contact Us

Version 1.9 Build 1488

News	FAQ
Search	Home

image.sepconvolve - Function

	Package	general
	Module	images
	Tool	image

Separable convolution

Synopsis
sepconvolve(outfile, axes, types, widths, scale, region, mask, overwrite, async)

Arguments

outfile	in	Output image file name
		Allowed:	String
		Default:	unset
axes	in	Axes to convolve
		Allowed:	Vector of integers
		Default:	[1,2,...]
types	in	Type of convolution kernel
		Allowed:	Vector of strings from `boxcar', `gaussian', and `hanning'
		Default:	'gaussian'
widths	in	Convolution kernel widths
		Allowed:	Vector of numeric, quantity or string
scale	in	Scale factor
		Allowed:	Float
		Default:	Autoscale
region	in	Region of interest
		Allowed:	Region tool
		Default:	Whole image
mask	in	OTF mask
		Allowed:	Boolean LEL expression or mask region
		Default:	None
overwrite	in	Overwrite (unprompted) pre-existing output file ?
		Allowed:	T or F
		Default:	F
async	in	Run asynchronously?
		Allowed:	Bool
		Default:	!dowait

Returns
Image tool or fail

Description

This function (short-hand name sc) does Fourier-based convolution of an image file by a specified separable kernel.

If outfile is given, the image is written to the specified disk file. If outfile is unset, the Image tool is associated with a temporary image. This temporary image may be in memory or on disk, depending on its size. When you destroy the Image tool (with the done function) this temporary image is deleted.

You specify which axes of the image you wish to convolve, by what kernel of what width. The kernel types can be shortened to `gauss', `hann' and `box'.

You specify the widths of the convolution kernels via the argument widths. The values can be specified as a vector of three different types.

• Quantity - for example widths=dq.quantity("1arcsec 0.00001rad"). Note that you can use pixel units, viz. widths=dq.quantity("10pix 0.00001rad") see below.

• String - for example widths="1km 2arcsec" (i.e. a string that the Quanta Constructor accepts).
• Numeric - for example widths=[10,20]. In this case, the units of the widths are assumed to be in pixels.

The interpretation of widths depends upon the kernel type.

• Gaussian - the specified width is the full-width at half-maximum.

• Boxcar (tophat) - the specified width is the full width.

• Hanning - The kernel is z[i] = 0.25*y[i - 1] + 0.5*y[i] + 0.25*y[i + 1]. The width is always 3 pixels, regardless of what you give (but you still have to give it !).

The scaling of the output image is determined by the argument scale. If you leave it unset, then autoscaling will be invoked which means that the convolution kernels will all be normalized to have unit volume to as to conserve flux.

If you do not leave scale unset, then the convolution kernel will be scaled by this value (it has peak unity before application of this scale factor).

Masked pixels will be assigned the value 0.0 before convolution. The output mask is the combination (logical OR) of the default input pixel mask (if any) and the OTF mask. Any other input pixel masks will not be copied. Use function maskhandler if you need to copy other masks too.

See also the other convolution functions convolve2d, convolve and hanning.

Example

- dowait := T
- im := image('xyv')
- im2 := im.sepconvolve(outfile='xyv.con', axes=[2,3], types="gauss box", widths=[10,20])