vhoughl − line Hough transform

Bvhoughl if=infile of=outfile [de=D] [th=T] [-i]

Vhoughl computes both the line Hough transform and the inverse line Hough transformation of infile. To compute the line Hough transformaion, infile has to be a two channel byte image. The first channel is interpreted as edge strength, the second as edge direction (which is given as an angle). The sobel edge detector, if invoked with the −d option, generates an appropriate file.

Outfile is again a byte image. Though infile can be an arbitrary xsize * ysize image, outfile will always be a size * size image, with size being the maximum of xsize and ysize.

The Hough space has coordinates r and p. An edge element (x, y; d) at position (x, y) in the input image and with direction d defines a unique line L, which contains (x, y) and whose normal vector points into direction d. If r denotes the distance between L and the origin, the above edge element gets mapped into the point (r, d) in Hough space.

If the -i flag is specified, vhoughl computes the inverse line Hough transform of infile, that is, for each nonzero pixel (r, p) in infile a line L is drawn into outfile, which has distance r from the origin and whose normal vector encloses angle p with the x-axis.

To "draw a line into outfile" means, that the value of each pixel on the line is incremented by one, unless it is already 255.

Infile and outfile are byte images and outfile has the same size as infile. Moreover, it hardly makes sense to apply vhoughl -i to any infile which has not been created by the line Hough transform.

Vhoughl works only with two channel byte images. Also, vhoughl -i requires infile to have square dimensions.

de=D

If a delta-d D is specified, each edge element gets mapped into a piece of a sine wave in hough space, whose length depends on D. Delta-d is not used when performing the inverse line Hough transform.

th=T

If a threshold T is specified, only edge elements with strength >= T are mapped to outfile. T defaults to 1.

-i

Specifies that the inverse line Hough transform should be used.

vhoughg, vhoughc,