lin-epics-modules/ADAndor
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
Files
0c51c857255a37fa91b65d48bb2afec927bfae3e
ADAndor/documentation/NDPluginStats.html

831 lines
23 KiB
HTML
Executable File
Raw Blame History

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xml:lang="en" xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>areaDetector Plugin NDPluginStats</title>
<meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type" />
</head>
<body>
<div style="text-align: center">
<h1>
areaDetector Plugin NDPluginStats</h1>
<h2>
March 20, 2010</h2>
<h2>
Mark Rivers</h2>
<h2>
University of Chicago</h2>
</div>
<h2>
Contents</h2>
<ul>
<li><a href="#Overview">Overview</a></li>
<li><a href="#Configuration">Configuration</a></li>
<li><a href="#Screens">Screen shots</a></li>
</ul>
<h2 id="Overview">
Overview
</h2>
<p>
NDPluginStats computes the following.
</p>
<ol>
<li>Basic statistics: minimum, maximum, mean, sigma, total, and net (background subtracted).</li>
<li>Centroid and sigma values in the X and Y dimensions.</li>
<li>Profiles of the array in the X and Y dimensions. A total of 8 profiles are calculated:
<ul>
<li>The average profiles in the X and Y directions. </li>
<li>The average profiles in the X and Y directions, for array elements greater than
the centroid threshold.</li>
<li>The profiles in the X and Y directions at the X and Y centroid position.</li>
<li>The profiles in the X and Y directions at a user-defined X and Y cursor position.</li>
</ul>
</li>
<li>A histogram of the values (e.g. number of pixels versus intensity per pixel).</li>
</ol>
<p>
Each calculcation can independently enabled and disabled. Calculations 1 and 4 can
be perfomed on arrays of any dimension. Calculations 2 and 3 are restricted to 2-D
arrays.
</p>
<p>
NDPluginStats inherits from NDPluginDriver. The <a href="areaDetectorDoxygenHTML/class_n_d_plugin_stats.html">
NDPluginStats class documentation</a> describes this class in detail.
</p>
<p>
NDPluginStats.h defines the following parameters. It also implements all of the
standard plugin parameters from <a href="pluginDoc.html#NDPluginDriver">NDPluginDriver</a>.
The EPICS database NDStats.template provide access to these parameters, listed in
the following table. Note that to reduce the width of this table the parameter index
variable names have been split into 2 lines, but these are just a single name, for
example <code>NDPluginStatsComputeStatistics</code>.
</p>
<table border="1" cellpadding="2" cellspacing="2" style="text-align: left">
<tbody>
<tr>
<td align="center" colspan="7,">
<b>Parameter Definitions in NDPluginStats.h and EPICS Record Definitions in NDStats.template</b></td>
</tr>
<tr>
<th>
Parameter index variable</th>
<th>
asyn interface</th>
<th>
Access</th>
<th>
Description</th>
<th>
drvInfo string</th>
<th>
EPICS record name</th>
<th>
EPICS record type</th>
</tr>
<tr>
<td align="center" colspan="7,">
<b>Basic statistics</b></td>
</tr>
<tr>
<td>
NDPluginStats<br />
ComputeStatistics</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Flag to control whether to compute statistics for this array (0=No, 1=Yes). Not
computing statistics reduces CPU load. Basic statistics computations are quite fast,
since they involve mostly double precision addition, with 1 multiply to compute
sigma, per array element.</td>
<td>
COMPUTE_STATISTICS</td>
<td>
$(P)$(R)ComputeStatistics<br />
$(P)$(R)ComputeStatistics_RBV</td>
<td>
bo<br />
bi</td>
</tr>
<tr>
<td>
NDPluginStats<br />
BgdWidth</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Width of the background in pixels to use when computing net counts. 0=no background
subtraction, so the net counts is the same as the total counts.</td>
<td>
BGD_WIDTH</td>
<td>
$(P)$(R)BgdWidth<br />
$(P)$(R)BgdWidth_RBV</td>
<td>
longout<br />
longin</td>
</tr>
<tr>
<td>
NDPluginStats<br />
MinValue</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Minimum value in any element in the array</td>
<td>
MIN_VALUE</td>
<td>
$(P)$(R)MinValue_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
MaxValue</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Maximum value in any element in the array</td>
<td>
MAX_VALUE</td>
<td>
$(P)$(R)MaxValue_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
MeanValue</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Mean value in the array</td>
<td>
MEAN_VALUE</td>
<td>
$(P)$(R)MeanValue_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
Total</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Sum (total) of all elements in the array. This is available as an ai record. The
total counts are also available as epicsInt32 values in an mca record via callbacks
to the drvFastSweep driver. The mca record is very useful for on-the-fly data acquisition
of the total counts in the detector or in an ROI.</td>
<td>
TOTAL</td>
<td>
$(P)$(R)Total_RBV<br />
$(P)$(R)TotalArray</td>
<td>
ai<br />
mca</td>
</tr>
<tr>
<td>
NDPluginStats<br />
Net</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Net (background subtracted) total of all elements in the array. The background is
calculated by determining the average counts per array element in a border around
the array of width NDPluginStatsBgdWidth. This average background counts per element
is then subtracted from all elements inside the array. If NDPluginStatsBgdWidth
is &le; 0 then no background is computed. The net counts is available as an ai record.
The net counts is also available as epicsInt32 values in an mca record via callbacks
to the drvFastSweep driver. The mca record is very useful for on-the-fly data acquisition
of the net counts in the detector or in an ROI.
</td>
<td>
NET</td>
<td>
$(P)$(R)Net_RBV<br />
$(P)$(R)NetArray</td>
<td>
ai<br />
mca</td>
</tr>
<tr>
<td>
NDPluginStats<br />
SigmaValue</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Sigma (standard deviation) of all elements in the array</td>
<td>
SIGMA_VALUE</td>
<td>
$(P)$(R)Sigma_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td align="center" colspan="7,">
<b>Centroid statistics</b></td>
</tr>
<tr>
<td>
NDPluginStats<br />
ComputeCentroid</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Flag to control whether to compute the centroid statistics (0=No, 1=Yes). The centroids
are computed from the average row and column profiles above the centroid threshold.
These calculations are also quite fast, since they just involve addition operations
for each array element.</td>
<td>
COMPUTE_CENTROID</td>
<td>
$(P)$(R)ComputeCentroid<br />
$(P)$(R)ComputeCentroid_RBV</td>
<td>
bo<br />
bi</td>
</tr>
<tr>
<td>
NDPluginStats<br />
CentroidThreshold</td>
<td>
asynFloat64</td>
<td>
r/w</td>
<td>
Threshold used when computing the centroid statistics. All array elements less than
this value are set to 0 for computing the centroid statistics. It is important to
set this value to ignore the "background" when computing the position and size of
a "beam" image, for example.</td>
<td>
CENTROID_THRESHOLD</td>
<td>
$(P)$(R)CentroidThreshold<br />
$(P)$(R)CentroidThreshold_RBV</td>
<td>
ao<br />
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
CentroidX</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
X centroid of the array above the centroid threshold.</td>
<td>
CENTROIDX_VALUE</td>
<td>
$(P)$(R)CentroidX_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
CentroidY</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Y centroid of the array above the centroid threshold.</td>
<td>
CENTROIDY_VALUE</td>
<td>
$(P)$(R)CentroidY_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
SigmaX</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Sigma X (width) of the distribution above the centroid threshold.</td>
<td>
SIGMAX_VALUE</td>
<td>
$(P)$(R)SigmaX_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
SigmaY</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Sigma Y (height) of the distribution above the centroid threshold.</td>
<td>
SIGMAY_VALUE</td>
<td>
$(P)$(R)SigmaY_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
SigmaXY</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
This is the normalized value of sigmaXY, i.e. sigmaXY/(sigmaX * sigmaY). This is
often called the correlation coefficient, r. It is zero if the X and Y profiles
are not correlated, meaning that the distribution is not tilted with respect to
the X and Y axes.</td>
<td>
SIGMAXY_VALUE</td>
<td>
$(P)$(R)SigmaXY_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td align="center" colspan="7,">
<b>X and Y Profiles</b></td>
</tr>
<tr>
<td>
NDPluginStats<br />
ComputeProfiles</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Flag to control whether to compute the profiles for this array (0=No, 1=Yes).</td>
<td>
COMPUTE_PROFILES</td>
<td>
$(P)$(R)ComputeProfiles<br />
$(P)$(R)ComputeProfiles_RBV</td>
<td>
bo<br />
bi</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileSizeX</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Number of array elements in the X profiles.</td>
<td>
PROFILE_SIZE_X</td>
<td>
$(P)$(R)ProfileSizeX_RBV</td>
<td>
longin</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileSizeY</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Number of array elements in the Y profiles.</td>
<td>
PROFILE_SIZE_Y</td>
<td>
$(P)$(R)ProfileSizeY_RBV</td>
<td>
longin</td>
</tr>
<tr>
<td>
NDPluginStats<br />
CursorX</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
X position of a user-defined cursor for profiles.</td>
<td>
CURSOR_X</td>
<td>
$(P)$(R)CursorX<br />
$(P)$(R)CursorX_RBV</td>
<td>
longout<br />
longin</td>
</tr>
<tr>
<td>
NDPluginStats<br />
CursorY</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Y position of a user-defined cursor for profiles.</td>
<td>
CURSOR_Y</td>
<td>
$(P)$(R)CursorY<br />
$(P)$(R)CursorY_RBV</td>
<td>
longout<br />
longin</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileAverageX</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
Profile of the average row in the array, i.e. the sum of all rows in the array divided
by the number of rows.</td>
<td>
PROFILE_AVERAGE_X</td>
<td>
$(P)$(R)ProfileAverageX_RBV</td>
<td>
waveform</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileAverageY</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
Profile of the average column in the array, i.e. the sum of all columns in the array
divided by the number of columns.</td>
<td>
PROFILE_AVERAGE_Y</td>
<td>
$(P)$(R)ProfileAverageY_RBV</td>
<td>
waveform</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileThresholdX</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
Same as ProfileAverageX except that all array elements less than CentroidThreshold
are set to zero when computing the average.</td>
<td>
PROFILE_THRESHOLD_X</td>
<td>
$(P)$(R)ProfileThresholdX_RBV</td>
<td>
waveform</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileThresholdY</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
Same as ProfileAverageY except that all array elements less than CentroidThreshold
are set to zero when computing the average.</td>
<td>
PROFILE_THRESHOLD_Y</td>
<td>
$(P)$(R)ProfileThresholdY_RBV</td>
<td>
waveform</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileCentroidX</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
X profile through the array in the row defined by CentroidY.</td>
<td>
PROFILE_CENTROID_X</td>
<td>
$(P)$(R)ProfileCentroidX_RBV</td>
<td>
waveform</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileCentroidY</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
Y profile through the array in the column defined by CentroidX.</td>
<td>
PROFILE_CENTROID_Y</td>
<td>
$(P)$(R)ProfileCentroidY_RBV</td>
<td>
waveform</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileCursorX</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
X profile through the array in the row defined by CursorY.</td>
<td>
PROFILE_CURSOR_X</td>
<td>
$(P)$(R)ProfileCursorX_RBV</td>
<td>
waveform</td>
</tr>
<tr>
<td>
NDPluginStats<br />
ProfileCursorY</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
Y profile through the array in the row defined by CursorX.</td>
<td>
PROFILE_CURSOR_Y</td>
<td>
$(P)$(R)ProfileCursorY_RBV</td>
<td>
waveform</td>
</tr>
<tr>
<td align="center" colspan="7,">
<b>Array histogram</b></td>
</tr>
<tr>
<td>
NDPluginStats<br />
ComputeHistogram</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Flag to control whether to compute the histogram for this array (0=No, 1=Yes). Not
computing the histogram reduces CPU load.</td>
<td>
COMPUTE_HISTOGRAM</td>
<td>
$(P)$(R)ComputeHistogram<br />
$(P)$(R)ComputeHistogram_RBV</td>
<td>
bo<br />
bi</td>
</tr>
<tr>
<td>
NDPluginStats<br />
HistSize</td>
<td>
asynInt32</td>
<td>
r/w</td>
<td>
Number of elements (bins) in the histogram</td>
<td>
HIST_SIZE</td>
<td>
$(P)$(R)HistSize<br />
$(P)$(R)HistSize_RBV</td>
<td>
longout<br />
longin</td>
</tr>
<tr>
<td>
NDPluginStats<br />
HistMin</td>
<td>
asynFloat64</td>
<td>
r/w</td>
<td>
Minimum value for the histogram. All values less than or equal to this will be in
the first bin of the histogram.</td>
<td>
HIST_MIN</td>
<td>
$(P)$(R)HistMin<br />
$(P)$(R)HistMin_RBV</td>
<td>
ao<br />
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
HistMax</td>
<td>
asynFloat64</td>
<td>
r/w</td>
<td>
Maximum value for the histogram. All values greater than or equal to this will be
in the last bin of the histogram.</td>
<td>
HIST_MAX</td>
<td>
$(P)$(R)HistMax<br />
$(P)$(R)HistMax_RBV</td>
<td>
ao<br />
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
HistEntropy</td>
<td>
asynFloat64</td>
<td>
r/o</td>
<td>
Entropy of the image. This is a measure of the sharpness of the histogram, and is
often a useful figure of merit for determining sharpness of focus, etc. It is defined
as -SUM(BIN[i]*log(BIN[i]), where the sum is over the number of bins in the histogram
and BIN[i] is the number of elements in bin i.</td>
<td>
HIST_ENTROPY</td>
<td>
$(P)$(R)HistEntropy_RBV</td>
<td>
ai</td>
</tr>
<tr>
<td>
NDPluginStats<br />
HistArray</td>
<td>
asynFloat64Array</td>
<td>
r/o</td>
<td>
Histogram array, i.e. counts in each histogram bin.</td>
<td>
HIST_ARRAY</td>
<td>
$(P)$(R)Histogram_RBV</td>
<td>
waveform</td>
</tr>
</tbody>
</table>
<p>
If the values of CentroidThreshold, CursorX, or CursorY are changed then the centroid
and profile calculations are performed again immediately on the last array collected.
Thus updated centroid statistics and profiles can be displayed even when new arrays
are not being acquired. These calculations are only performed when enabled by ComputeCentroid
and ComputeProfiles.</p>
<h2 id="Configuration">
Configuration</h2>
<p>
The NDPluginStats plugin is created with the NDStatsConfigure command, either from
C/C++ or from the EPICS IOC shell.</p>
<pre>NDStatsConfigure(const char *portName, int queueSize, int blockingCallbacks,
const char *NDArrayPort, int NDArrayAddr,
int maxBuffers, size_t maxMemory,
int priority, int stackSize)
</pre>
<p>
For details on the meaning of the parameters to this function refer to the detailed
documentation on the NDStatsConfigure function in the <a href="areaDetectorDoxygenHTML/_n_d_plugin_stats_8cpp.html">
NDPluginStats.cpp documentation</a> and in the documentation for the constructor
for the <a href="areaDetectorDoxygenHTML/class_n_d_plugin_stats.html">NDPluginStats
class</a>.
</p>
<h2 id="Screens">
Screen shots</h2>
<p>
The following MEDM screen provides access to the parameters in NDPluginDriver.h
and NDPluginStats.h through records in NDPluginBase.template and NDStats.template.
</p>
<div style="text-align: center">
<h3>
NDStats.adl</h3>
<img alt="NDStats.png" src="NDStats.png" />
</div>
<p>
The following MEDM screen shows the average profile of an image in the X direction.
</p>
<div style="text-align: center">
<h3>
NDPlot.adl</h3>
<img alt="NDStats_AverageX.png" src="NDStats_AverageX.png" />
</div>
<p>
The following MEDM screen shows the profile of an image in the Y direction at the
location of the user-defined cursor.
</p>
<div style="text-align: center">
<h3>
NDPlot.adl</h3>
<img alt="NDStats_CursorY.png" src="NDStats_CursorY.png" />
</div>
<p>
The following MEDM screen shows the histogram of intensities of an array.
</p>
<div style="text-align: center">
<h3>
NDPlot.adl</h3>
<img alt="NDStats_Histogram.png" src="NDStats_Histogram.png" />
</div>
<p>
The following MEDM screen combines many parameters for 5 NDPluginStats plugins on
a single screen.
</p>
<div style="text-align: center">
<h3>
NDStats5.adl</h3>
<img alt="NDStats5.png" src="NDStats5.png" />
</div>
<p>
The following MEDM screen shows the MCA record containing the total counts from
the Stats1 plugin. This is the total counts as a function of time, and is very useful
for on-the-fly data acquisition, where the NDStats plugin computes the net or total
counts in the detector or an ROI.
</p>
<div style="text-align: center">
<h3>
mca.adl</h3>
<img alt="NDStatsTotalArray.png" src="NDStatsTotalArray.png" />
</div>
<p>
The following MEDM screen shows the MCA record containing the net counts from the
Stats1 plugin.
</p>
<div style="text-align: center">
<h3>
mca.adl</h3>
<img alt="NDStatsNetArray.png" src="NDStatsNetArray.png" />
</div>
</body>
</html>
Reference in New Issue View Git Blame Copy Permalink