1Pamgauss User Manual(0) Pamgauss User Manual(0)
2
6 pamgauss - create a two-dimensional Gaussian function as a PAM image
7
10 pamgauss width height -sigma=number [-maxval=number] [-tuple‐
11 type=string] [-maximize] [-oversample=number]
12 Minimum unique abbreviation of option is acceptable. You may use dou‐
14 ble hyphens instead of single hyphen to denote options. You may use
15 white space in place of the equals sign to separate an option name from
16 its value.
17
21 pamgauss 7 7 -sigma=.5 -maximize -tupletype=GRAYSCALE | pamtopnm >gauss.pgm
22 pnmconvol -nooffset -normalize gauss.pgm myimage.ppm >blurred.ppm
23
27 This program is part of Netpbm(1).
28 pamgauss generates a one-plane PAM image whose samples are a Gaussian
30 function of their distance from the center of the image. I.e. the
31 sample value is highest in the center and goes down, in a bell curve
32 shape, as you move away from the center.
33 You can use this image as a convolution kernel with pnmconvol to blur
35 an image. (This technique is known as Gaussian blurring).
36 width and height are the dimensions of the image that pamgauss gener‐
38 ates. Mathematically speaking, they are the domain of the two-dimen‐
39 sional Gaussian function. If you want to be sure you get a whole
40 Gaussian function, make sure that you choose a standard deviation and
41 image dimensions so that if you made it any larger, the sample values
42 at the edges would be zero.
43 The output image is PAM. To make it usable with pnmconvol, specify
45 -tupletype=GRAYSCALE so pnmconvol can use it as if it were PGM. You
46 must use the -nooffset option on pnmconvol because zero means zero in
47 the PAM that pamgauss generates.
48 Without -maximize, the sum of all the samples is equal to the image's
50 maxval (within rounding error). This is true even if you clip the
51 Gaussian function by making the image too small. This is what is nor‐
52 mally required of a convolution kernel.
53 pamgauss oversamples and averages to represent the continuous Gaussian
55 function in discrete samples in the PAM output. Consider an image 11
56 samples wide and an oversampling factor of 10. The samples can be
57 thought of as contiguous squares one unit wide. The center of the im‐
58 age is thus the center of the 6th sample from the left. The 3rd sample
59 from the left covers a range of distances from 3 to 4 units from the
60 center of the image. Because the oversampling factor is 10, pamgauss
61 computes the value of the Gaussian function at 10 points evenly spaced
62 between 3 and 4 units from the center of the image and assigns the 3rd
63 sample from the left the mean of those 10 values.
64
68 In addition to the options common to all programs based on libnetpbm
69 (most notably -quiet, see
70 Common Options ⟨index.html#commonoptions⟩ ), pamgauss recognizes the
71 following command line options:
72 -sigma=number
76 This is the standard deviation of the Gaussian function. The
77 higher the number, the more spread out the function is. Nor‐
78 mally, you want to make this number low enough that the function
79 reaches zero value before the edge of your image.
80 number is in units of samples.
82 This option is required. There is no default.
84 -maximize
87 Causes pamgauss to use the whole dynamic range available in the
88 output PAM image by choosing an amplitude for the Gaussian func‐
89 tion that causes the maximum value in the image to be the maxval
90 of the image.
91 If you select this, you probably want to normalize the output
93 (scale the samples down so the volume under the surface of the
94 two-dimensional Gaussian function is the maxval) before you use
95 it, for example with pnmconvol's -normalize option. The reason
96 this is different from just not using -maximize is that this
97 subsequent normalization can be done with much more precision
98 than can be represented in a PAM image.
99 Without this option, pamgauss uses an amplitude that makes the
101 volume under the surface of the two-dimensional Gaussian func‐
102 tion the maxval of the image. This means all the samples in the
103 image are normally considerably less than the maxval.
104 This option was new in Netpbm 10.79 (June 2017).
106 -maxval=number
109 This is the maxval for the output image. 65535 is almost always
110 the best value to use. But there may be some programs (not part
111 of Netpbm) that can't handle a maxval greater than 255.
112 The default is 255.
114 -tupletype=string
117 This is the value of the "tuple_type" attribute of the created
118 PAM image. It can be any string up to 255 characters.
119 If you don't specify this, pamgauss generates a PAM with unspec‐
121 ified tuple type.
122 -oversample=number
125 This sets the oversampling factor. pamgauss samples the Gauss‐
126 ian function this many times, both horizontally and vertically,
127 to get the value of each sample in the output.
128 An oversampling factor of 1 means no oversampling, which means
130 each sample is based only on the value of the Gaussian function
131 at the center of the sample.
132 The default is 5 divided by the standard deviation, rounded up
134 to a whole number.
135 This option was new in Netpbm 10.79 (June 2017). Before that,
137 it is essentially 1 - there is no oversampling.
138
143 pnmconvol(1), pamtopnm(1), pgmkernel(1), pamseq(1), pam(1)
144
147 pamgauss was new in Netpbm 10.23 (July 2004).
148
150 This manual page was generated by the Netpbm tool 'makeman' from HTML
151 source. The master documentation is at
152 http://netpbm.sourceforge.net/doc/pamgauss.html
154netpbm documentation 18 May 2017 Pamgauss User Manual(0)