1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
|
.help imhistogram Nov89 images.imutil
.ih
NAME
imhistogram -- print or plot the histogram of an image
.ih
USAGE
imhistogram image
.ih
PARAMETERS
.ls image
The name of the image or image subsection whose histogram is to be calculated.
.le
.ls z1 = INDEF, z2 = INDEF
The minimum and maximum histogram intensity. The image minimum and maximum
pixel values are used by default.
.le
.ls binwidth = INDEF
The resolution of the histogram in counts. If \fIbinwidth\fR is not defined,
the parameter \fInbins\fR determines the histogram resolution.
.le
.ls nbins = 512
The number of bins in, or resolution of, the histogram.
The \fInbins\fR parameter is overridden if \fIbinwidth\fR is defined.
.le
.ls autoscale = yes
In the case of integer data, automatically adjust \fInbins\fR and
\fIz2\fR to avoid aliasing effects.
.le
.ls top_closed = no
Include z2 in the top bin? Each bin of the histogram is a subinterval
that is half open at the top. \fITop_closed\fR decides whether those
pixels with values equal to z2 are to be counted in the histogram. If
\fBtop_closed\fR is yes, the top bin will be larger than the other bins.
.le
.ls hist_type = "normal"
The type of histogram to plot or list. The choices are "normal",
"cumulative", "difference", or "second_difference". The two
"difference" options are calculated as forward differences, i.e.,
diff[n] = hist[n+1] - hist[n].
.le
.ls listout = no
List instead of plot the histogram? The list is never log scaled.
.le
.ls plot_type = "line"
The plot vector type. The options are "line" and "box".
.le
.ls logy = yes
Use log scaling on the y-axis of the plot?
.le
.ls device = "stdgraph"
The output graphics device.
.le
.ih
DESCRIPTION
\fIimhistogram\fR calculates the histogram of the IRAF image
\fIimage\fR using the parameters \fInbins\fR, \fIz1\fR and \fIz2\fR.
If either \fIz1\fR or \fIz2\fR is undefined the image minimum or
maximum is used. If \fIlistout\fR = no, the histogram is plotted on
the graphics device \fIdevice\fR in the vector mode specified by
\fIplot_type\fR. The plot may be log scaled if \fIlogy\fR = yes (the
default). If \fIlistout\fR = yes, the histogram is listed on the
standard output.
In addition to producing the "normal" histogram, the task will also
calculate cumulative and marginal (forward difference) histograms
depending on the choice of the \fIhist_type\fR parameter (choices
are: "normal", "cumulative", "difference", and "second_difference").
The plot will be labeled by the type of histogram as well as the image
name and title and the binning parameters.
Each bin of the histogram is defined to be half open at the top. This
results in an ambiguity deciding whether those pixels with z=z2 are
included in the topmost bin. This decision is left to the user via the
\fItop_closed\fR parameter. This is usually only important with integer
images and histograms with few bins.
.ih
EXAMPLES
1. Output the histogram of an image to a file.
cl> imhist M51.imh li+ nbins=100 > fits1.hst
2. Plot the histogram of another image between the values 0 and 2000.
cl> imhist M31.imh nbins=100 z1=0. z2=2000.
3. Ditto, but set the histogram resolution explicitly to avoid
smoothing the histogram.
cl> imhist M31.imh nbins=100 z1=0 z2=2000 nbins=2001
4. Plot the cumulative histogram. This is most useful for images with
fairly flat "normal" histograms.
cl> imhist R50.imh hist=cum
.ih
BUGS
If the resolution of the histogram (number of bins) is a non-integral multiple
of the intensity resolution of the data (number of possible intensity values),
then \fIaliasing\fR can occur. The effect is to cause periodic zero dropouts
(for an oversampled histogram) or excess-valued bins (for a slightly
undersampled histogram). The \fIautoscaling\fR feature, if enabled, will
adjust the histogram parameters to avoid such aliasing effects for integer
data. This is not possible for floating point data, however, in which case
aliasing is certainly possible and can only be avoided by manually adjusting
the histogram parameters. One should also be aware that \fIsmoothing\fR of
the histogram will occur whenever the data range exceeds the histogram
resolution.
.ih
SEE ALSO
listpixels, plot.graph, proto.mkhistogram
.endhelp
|
