V3EMBEDVisionX V4 NAME

v3embed − pad input image by zero padding or image embeding.

SYNOPSIS

v3embed if= of= [ig=] [-e] [x=] [y=] [z=] [-v] [-d]

DESCRIPTION

v3embed can increase the size of an image input if= in two ways: zero-padding or embedding in another (parent) image.

In the zero-padding mode, the output image of= is of the same image type of the input if=. The output bbx can be specified in two ways: the padding sizes in each dimension can be specifeid by x=, y= and z=; alternatively, if ig= is specified and -e is not specified, the output bbx is the same as the bbx of the input ig=. The values of pixels in if= that are within output bbx are preserved, and the rest pixels in the output of= are set to zero.

In the embeding mode (when ig= and -e are specifeid), the output image of= has the same image type and bbx as the input reference image ig=. The input if= is embeded into the reference ig= in two steps: The ig= pixel values are first copied to the output of=. Then the non-zero pixels in if= within the output bbx are copied to the output of=.

Note that the bbx of the input reference ig= is not required to include the bbx of if=, as long as the intersection of the boiunding bioxes is not empty. All image types (including byte, short, int, float and double) are accepted.

OPTIONS

	if=		input image to be padded or embeded into ig=.
	of=		output zero-padded or embeded image.
	ig=		input image as the size reference.
	-e		embedding mode, i.e., embed if= into ig=. Default zero-padding.
	x=		padding size in x direction (x or xl,xh).
	y=		padding size in y direction (y or yl,yh).
	z=		padding size in z direction (z or zl,zh).
	-v		verbose mode.
	-d		debug mode.

SEE ALSO

vembed(1)

AUTHOR

Shuang Liu, March 2016