1Pnmhisteq User Manual(0) Pnmhisteq User Manual(0)
2
6 pnmhisteq - histogram equalize a PNM image
7
10 pnmhisteq
11 [-gray]
13 [-rmap pgmfile]
15 [-wmap pgmfile]
17 [-verbose]
19 [pnmfile]
21
25 This program is part of Netpbm(1).
26 pnmhisteq increases the contrast of a PGM or PPM image through the
28 technique of 'histogram equalization.'[1]
29 pnmhisteq computes a histogram of the luminosity of the pixels in the
31 image. It then calculates a mapping between each luminosity and a new
32 luminosity such that it spreads out intensity levels around histogram
33 peaks and compresses them at troughs. I.e. it moves pixels around in
34 the histogram so as to make it flat. It applies that mapping to the
35 input image to produce the output image. The effect of this is that
36 the image has equal numbers of pixels at each possible intensity level,
37 which means it uses the available levels of intensity more efficiently
38 and thereby has more visible detail.
39 Mathematically, the luminosity mapping is this: Assume the pixels are
41 sorted by luminosity into B buckets numbered from 0 (lowest luminosity)
42 to B-1. N[i] is the number of pixels in bucket i. T is the total num‐
43 ber of pixels (sum of N[i] over all i). W is the luminosity of white.
44 pnmhisteq replaces an input pixel whose luminosity falls into bucket j
46 with one whose luminosity is:
47 j
50 ---
51 \
52 > (N[i] / T) * W
53 /
54 ---
55 i=0
56 Considering a grayscale image for simplicity, this means that pixels in
58 the most luminous bucket become white. Pixels in the 10th per centile
59 of luminosity become 10% of white.
60 pnmhisteq maps a single luminosity in the input to a single luminosity
62 in the output. That means if pixels A and B both have luminosity .2 in
63 the input, and pixel A has luminosity .4 in the output, pixel B also
64 has luminosity .4 in the output. And since the luminosities in the
65 input are not continuous, the luminosities in the output aren't either
66 and pnmhisteq doesn't meet the ideal of having exactly the same number
67 of pixels of each luminosity in the output.
68 If you're processing a related set of images, for example frames of an
70 animation, it's generally best to apply the same luminosity mapping to
71 every frame, since otherwise you'll get distracting frame-to-frame
72 changes in the brightness of objects. pnmhisteq's -wmap option allows
73 you to save, as a PGM image, the luminosity map it computes from an
74 image. The -rmap option causes pnmisteq to use such an image as its
75 luminosity map.
76 So you can run pnmhisteq with -wmap on a composite you created with
78 pnmcat of the images you intend to process. Then, you can run pnmisteq
79 with -rmap on each of the individual images, using the luminosity map
80 you generated from the composite.
81 Use pnmhistmap to see the result. Run a color image through ppmtopgm
83 first so that you see a histogram of the luminosity instead of his‐
84 tograms of the three color components. It should generally show a flat
85 histogram. But due to the quantization effects described above, you
86 might see high bars interleaved with low bars, with the local average
87 being flat. To see local averages, use the -width option of
88 pnmhistmap.
89
93 You can abbreviate any option to its shortest unique prefix.
94 -gray When processing a color image, only gray pixels (those with
98 identical red, green, and blue values) are included in the his‐
99 togram and modified in the output image. This is a special pur‐
100 pose option intended for images where the actual data are gray
101 scale, with color annotations you don't want modified. Weather
102 satellite images that show continent outlines in color are best
103 processed using this option. The option has no effect when the
104 input is a graymap.
105 -rmap mapfile
108 Process the image using the luminosity map specified by the PGM
109 file mapfile.
110 The PGM image, usually created by an earlier run of pnmhisteq
112 with the -wmap option, contains a single row with number of col‐
113 umns equal to the maxval (greatest intensity value) of the image
114 plus one. Each pixel in the image is transformed by looking up
115 its luminosity in the corresponding column in the map file (col‐
116 umn number = luminosity) and changing it to the value given by
117 that column.
118 -wmap mapfile
121 Creates a PGM file mapfile, containing the luminosity map com‐
122 puted from the histogram of the input image. This map file can
123 be read on subsequent runs of pnmhisteq with the -rmap option,
124 allowing a group of images to be processed with an identical
125 map.
126 -verbose
129 Prints the histogram and luminosity map on Standard Error.
130
135 Histogram equalization is effective for increasing the visible detail
136 in scientific imagery and in some continuous-tone pictures. It is
137 often too drastic, however, for scanned halftone images, where it does
138 an excellent job of making halftone artifacts apparent. You might want
139 to experiment with pnmnorm and pnmgamma for more subtle contrast
140 enhancement.
141 The luminosity map file supplied by the -rmap option must have the same
143 maxval as the input image. This is always the case when the map file
144 was created by the -wmap option of pnmhisteq. If this restriction
145 causes a problem, simply adjust the maxval of the map with pamdepth to
146 agree with the input image.
147 If the input is a PBM file (on which histogram equalization is an iden‐
149 tity operation), the only effect of passing the file through pnmhisteq
150 will be the passage of time.
151
154 pnmnorm(1), pnmcat(1), pamdepth(1), pnmgamma(1), pnm(1),
155 [1] Russ, John C. The Image Processing Handbook. Boca Raton: CRC
159 Press, 1992. Pages 105-110.
160netpbm documentation 02 February 2010 Pnmhisteq User Manual(0)