manpage oopses

ChangeLog

@@ -1,6 +1,7 @@
 25/3/09 started 7.18.0
 - revised version numbers
 - updated vipsmanual
+- revised manpages
 
 6/3/09 started 7.17.3
 - revised nohalo

TODO

@@ -1,7 +1,3 @@
-- im_rank.3 has a note about a black border, argh
-
-  any other references to this?
-
 - VipsFormat.3 man page should list all formats, since eg. radiance does not
   have a separate im_rad2vips page
 

man/im_compass.3

@@ -61,12 +61,12 @@ using look-up tables. The output is integer.  coefficients.  The size of the
 output image and the number of channels are the same as the corresponding of
 the input.
 
-If the sizes of the mask are xm and ym and the sizes of the image are xs and ys,
-then there is a black border at the output image as follows:
+The output image is the same size as the input. The edge pixels are calculated
+by expanding the input image using
+.B im_embed(3)
+in mode 1 (replicating edge pixels) just enough so that the output can match
+the input.
 
-The top ym/2 lines black, bottom ys-ym/2-ym lines black; each of the remaining
-lines has the initial xm/2 pels blank, and the final
-xs-xm/2-xm pels black (division over 2 is integer division).
 The output at each point is divided by scale and then the offset is added.
 Both offset and scale are held in mask.
 

man/im_conv_raw.3

@@ -1,123 +1 @@
-.TH CONVOLUTION 3 "2 May 1991"
-.SH NAME
-im_conv, im_convf, im_convsep, im_convsepf, 
-im_convsub, im_shrink \-  convolves an image with a generalised 
-mask
-.SH SYNOPSIS
-#include <vips/vips.h>
-
-int im_conv(in, out, mask)
-.br
-IMAGE *in, *out;
-.br
-INTMASK *mask;
-
-int im_convf(in, out, mask)
-.br
-IMAGE *in, *out;
-.br
-DOUBLEMASK *mask;
-
-int im_convsep(in, out, mask)
-.br
-IMAGE *in, *out;
-.br
-INTMASK *mask;
-
-int im_convsepf(in, out, mask)
-.br
-IMAGE *in, *out;
-.br
-DOUBLEMASK *mask;
-
-int im_convsub(in, out, mask, xskip, yskip)
-.br
-IMAGE *in, *out;
-.br
-INTMASK *mask;
-.br
-int xskip, yskip;
-
-int im_shrink(in, out, xfactor, yfactor)
-.br
-IMAGE *in, *out;
-.br
-double xfactor, yfactor;
-
-.SH DESCRIPTION
-These functions
-convolve the image pointed by the image descriptor in with the file
-pointed by mask and puts the result in the image pointed by out.
-The mask structures INTMASK and DOUBLEMASK are 
-returned by the function im_read_imask(3) or im_read_dmask(3),
-for integer and double masks respectively.
-
-Input should be non-complex.  
-The size and type of the output image are the same as the size of the input.  
-To
-output a larger type (for example, to output an int image from convolution
-of a byte image, avoiding clipping), cast the input image up with one of
-the im_clip2*() functions.
-
-If the sizes of the mask are xm and ym and the sizes of the image are xs and ys,
-then there is a black border at the output image as follows:
-
-The top ym/2 lines black, bottom ys-ym/2-ym lines black; each of the remaining
-lines has the initial xm/2 pels blank, and the final
-xs-xm/2-xm pels (division over 2 is integer division).
-
-The output at each point is divided by scale and then the offset is added.
-Both offset and scale are held in mask.
-
-.B im_conv()
-and
-.B im_convsep()
-have 'raw' versions which do not add the black border: instead the output
-image is smaller than the input.
-
-.B im_conv()
-and
-.B im_conv_raw()
-convolve any non-complex input image to make an output image of the same
-type. Rounding is appropriate to the image type.
-
-.B im_convf()
-convolves to float (double if the input is double).
-The function expects a double mask.
-
-.B im_convsep()
-and 
-.B im_convsep_raw()
-carry out convolution using an 1xN or Nx1 separable mask.  
-
-The function
-scales the output result by dividing it with scale*scale. The scale
-factor should therefore be the sqrt of the scale of the square NxN mask.
-
-Rounding is appropriate to the image type.  It works on any non-complex image,
-and writes the output in the same format as the input.
-The function expects integer mask.
-
-.B im_convsepf()
-convolves to float (double if the input is double).
-The function expects a double mask.
-
-.B im_convsub()
-convolves the byte image pointed by in and writes the result as a byte output.
-Using this function the input image is subsampled on
-both directions by an integer factor of xskip horizontally and
-an integer factor of yskip vertically.
-During the covolution, values are rounded before division.
-Both input and output are bytes.  Output is clipped between 0 and 255.
-The function expects an integer mask.
-
-.B im_shrink()
-shrink the input image file by xfactor along the horizontal and
-yfactor along the vertical direction.  The function doesnot perform subpixel
-interpolation and therefore the resultant image can present aliasing especially
-for small x and y factors. Any size image, any non-complex type, any number of
-bands.
-.SH RETURN VALUE
-The function returns 0 on success and -1 on error.
-.SH SEE ALSO
-im_read_imask(3), im_compass(3), im_fastcor(3).
+.so man3/im_conv.3

man/im_dilate.3

@@ -40,33 +40,42 @@ Mask coefficients can be either 0 (for object) or 255 (for background) or 128
 The mask should have odd length sides and the origin of the mask is at location
 (m-\>xsize/2,m-\>ysize/2) integer division.  All algorithms have been based on
 the book "Fundamentals of Digital Image Processing" by A.  Jain, pp 384-388,
-Prentice-Hall, 1989.  Essentially, im_dilate(3) sets pixels in the output if
-*any* part of the mask matches, whereas im_erode(3) sets pixels only if *all*
+Prentice-Hall, 1989.  Essentially, 
+.B im_dilate(3) 
+sets pixels in the output if
+*any* part of the mask matches, whereas 
+.B im_erode(3) 
+sets pixels only if *all*
 of the mask matches.
 
-im_dilate(3)
+.B im_dilate(3)
 dilates the image pointed by in, according to the mask pointed by m and writes
 the result in the location pointed by the IMAGE descriptor out. The output
-image is the same size as the input, with a black border in the manner of
-im_conv(3).
+image is the same size as the input, in the manner of
+.B im_conv(3).
 
-im_dilate_raw(3)
-works as im_dilate(3), but does not add a black border.
+.B im_dilate_raw(3)
+works as im_dilate(3), but does not expand the input.
 
-im_erode(3)
+.B im_erode(3)
 erodes the image pointed by in, according to the mask pointed by m and writes
 the result in the location pointed by the IMAGE descriptor out. Again, the
 output image is forced to have the same size as the input.
 
-im_erode_raw(3)
-works as im_erode(3), but does not add a black border.
+.B im_erode_raw(3)
+works as im_erode(3), but does not expand the input.
 
-See the boolean operations im_and(3), im_or(3) and im_eor(3) for analogues
+See the boolean operations 
+.B im_andimage(3), 
+.B im_orimage(3) 
+and 
+.B im_eorimage(3) 
+for analogues
 of the usual set difference and set union operations.
 
 .SH RETURN VALUE
 All functions returns 0 on success and -1 on error.
 .SH SEE ALSO
-im_read_imask(3), im_conv(3), im_and(3), im_rotate_imask(3).
+im_read_imask(3), im_conv(3), im_andimage(3), im_rotate_imask(3).
 .SH COPYRIGHT
 1991-1995, Birkbeck College and the National Gallery

man/im_heq.3

@@ -21,28 +21,32 @@ int im_hsp( in, ref, out )
 IMAGE *in, *ref, *out;
 
 .SH DESCRIPTION
-im_heq()
+.B im_heq(3)
 histogram equalises the unsigned char image held by the IMAGE descriptor
 in.  The result is written to the IMAGE descriptor out.
 If bandno is -1 then all input bands are equalised independently.  In all
 other cases the input image is equalised using the histogram of bandno only.
 The latter processing produces better results.
 
-im_hsp()
+.B im_hsp(3)
 maps in to out with histogram specified by the ref.  All images should be
 unsigned char.  Each band of the output image is specified according to the
 distribution of grey levels of the reference image according to
 im_histspec(3).
 
-im_lhisteq()
+.B im_lhisteq(3)
 histogram equalises the one channel unsigned char image pointed to by the
 Image descriptor in.  The result is written to the IMAGE descriptor out.
 The histogram equalisation is based on a window of size xw by yw centered at
-the current location of each input pixel.  The produced image has a black
-border of sizes xw/2 (left), xw-xw/2 (right), yw/2 (top) and yw-yw/2 (bottom).
+the current location of each input pixel. 
 
-im_lhisteq_raw()
-is as above, but does not add the black border to the output image.
+In the manner of 
+.B im_conv(3)
+the input image is expanded so that the output image is the same size as the
+input.
+
+.B im_lhisteq_raw(3)
+is as above, but does not expand the input.
 
 .SH RETURN VALUE
 All functions returns 0 on success and -1 on error.

man/im_rank.3

@@ -23,12 +23,14 @@ is passed over the image. At each position, the pixels inside the window are
 sorted into ascending order and the pixel at the nth position is output. n
 numbers from 0.
 
-It works for any non-complex image type, with any number of bands.  A black
-border is added to the output image to make it the same size as the
-input.
+It works for any non-complex image type, with any number of bands. 
+The input
+is expanded by copying edge pixels before performing the operation so that the
+output image has the same size as the input. Edge pixels in the output image
+are therefore only approximate.
 
 .B im_rank_raw()
-works just as im_rank(), but does not add the border.
+works just as im_rank(), but does not expand the input.
 .SH EXAMPLES
 For a median filter with mask size m (3 for 3x3, 5 for 5x5, etc.) use