diff options
Diffstat (limited to 'doc/profcheck.html')
| -rw-r--r-- | doc/profcheck.html | 285 |
1 files changed, 165 insertions, 120 deletions
diff --git a/doc/profcheck.html b/doc/profcheck.html index ef0bc5b..a85e07e 100644 --- a/doc/profcheck.html +++ b/doc/profcheck.html @@ -3,7 +3,7 @@ <head> <title>profcheck</title> <meta http-equiv="content-type" content="text/html; - charset=ISO-8859-1"> + charset=windows-1252"> <meta name="author" content="Graeme Gill"> </head> <body> @@ -11,22 +11,25 @@ <h3>Summary</h3> Check an <a href="File_Formats.html#ICC">ICC</a> profile against <a href="File_Formats.html#.ti3">.ti3</a> test chart - data.<br> + data, and optionally create a pruned .ti3 file.<br> <h3>Usage Summary</h3> <small><span style="font-family: monospace;">profcheck - [-options] - data.ti3 iccprofile.icm</span><br style="font-family: + [-options] data.ti3 iccprofile.icm</span><br style="font-family: monospace;"> - - <span style="font-family: monospace;"></span></small><small><span + <span style="font-family: monospace;"></span></small><small><span style="font-family: monospace;">-v - [level] - Verbosity level (default 1), 2 to print each DE</span></small><small><span + [level] Verbosity level (default + 1), 2 to print each DE</span></small><small><span style="font-family: monospace;"></span><br style="font-family: monospace;"> <span style="font-family: monospace;"> -c Show + + + + + CIE94 delta E values</span><br style="font-family: monospace;"> <span style="font-family: monospace;"> -k @@ -35,157 +38,204 @@ Show <span style="font-family: monospace;"> -w create - VRML visualization (iccprofile.wrl)</span><br + + + + + + X3DOM visualization (iccprofile.x3d.html)</span><br style="font-family: monospace;"> <span style="font-family: monospace;"> -x Use - VRML axes<br> + + + + + + X3DOM axes<br> -m Make -VRML +X3DOM + + + + + lines a minimum of 0.5<br style="font-family: monospace;"> </span><span style="font-family: monospace;"> -e Color vectors - acording to delta E</span><br style="font-family: monospace;"> - <span style="font-family: monospace;"> -d + + + + + + acording to delta E<br> + -s + Sort output by delta E<br> + -h + + + + + Plot a histogram of delta E's<br> + -P + de + Create a pruned .ti3 with points less or equal to de delta E<br + style="font-family: monospace;"> + </span> <span style="font-family: monospace;"> -d devval1,deval2,devvalN</span><br style="font-family: monospace;"> <span style="font-family: monospace;"> Specify - a device value to sort - against</span><br style="font-family: monospace;"> + + + + + + a device value to sort against</span><br style="font-family: + monospace;"> <span style="font-family: monospace;"> -p Sort + + + + + device value by PCS/Lab target</span><br style="font-family: monospace;"> - - <span style="font-family: monospace;">-f + <span style="font-family: monospace;">-f [illum] Use Fluorescent Whitening Agent compensation [opt. simulated inst. illum.:<br> + + + + + M0, M1, M2, A, C, D50 (def.), D50M2, D65, F5, F8, F10 or file.sp]<br> -i illum Choose illuminant for computation of CIE XYZ from spectral data & FWA:<br> + + + + + A, C, D50 (def.), D50M2, D65, F5, F8, F10 or file.sp</span><span style="font-family: monospace;"></span><br style="font-family: monospace;"> <span style="font-family: monospace;"> -o - observ - Choose CIE - Observer - for spectral data:</span><br style="font-family: monospace;"> + observ Choose CIE Observer for + spectral data:</span><br style="font-family: monospace;"> <span style="font-family: monospace;"></span></small><small><span style="font-family: monospace;"> - 1931_2 - </span></small><small><span style="font-family: monospace;">(def.)</span></small><small><span - style="font-family: monospace;">, - 1964_10, S&B - 1955_2, shaw, J&V 1978_2</span></small><small><span - style="font-family: monospace;"><br> + + + + + + 1931_2 </span></small><small><span style="font-family: + monospace;">(def.)</span></small><small><span + style="font-family: monospace;">, 1964_10, S&B 1955_2, shaw, + J&V 1978_2</span></small><small><span style="font-family: + monospace;"><br> -I intent r = relative colorimetric, a = absolute (default)<br style="font-family: monospace;"> </span><i style="font-family: monospace;"> data.ti3</i><span style="font-family: monospace;"> - - Test - point data file</span><br style="font-family: monospace;"> + Test point data file</span><br style="font-family: + monospace;"> <i style="font-family: monospace;"> iccprofile.icm</i><span style="font-family: monospace;"> Profile to check</span></small> <br> <h3>Usage Details and Discussion</h3> <b> profcheck</b> provides a way of checking how well an <a href="File_Formats.html#ICC">ICC</a> profile conforms to the test - sample data that was used to create it (or other test samples - that are from the same device). This is the same sort of check done - within - the profile making tool (<a href="colprof.html">colprof</a>), but - having a - separate tool provides some flexibility. The absolute forward - table in the profile is used to create PCS values from the sample - points, and the profiles PCS value then compared to the PCS values - of - the measured sample points. Note the lower delta E values are not - always a better measure of how good a profile is. The aim of a - profile - is to model the underlying characteristics of a device, not to - slavishly reproduce the sampled data point values. Sampled data - point - values contain device variation and instrument reading inaccuracies, - and a good profiler will try and filter out this noise, resulting in - some deliberate differences between the profile and the sample - points - used to create it.<br> + sample data that was used to create it (or other test samples that + are from the same device). This is the same sort of check done + within the profile making tool (<a href="colprof.html">colprof</a>), + but having a separate tool provides some flexibility. The + absolute forward table in the profile is used to create PCS values + from the sample points, and the profiles PCS value then compared to + the PCS values of the measured sample points. Note the lower delta E + values are not always a better measure of how good a profile is. The + aim of a profile is to model the underlying characteristics of a + device, not to slavishly reproduce the sampled data point values. + Sampled data point values contain device variation and instrument + reading inaccuracies, and a good profiler will try and filter out + this noise, resulting in some deliberate differences between the + profile and the sample points used to create it.<br> <br> The <b>-v</b> flag prints out extra information during the - checking. A - value greater than 1 will print the color values of each test point.<br> + checking. A value greater than 1 will print the color values of each + test point.<br> <br> The <b>-c</b> option causes the differences between the test values - and - the profile prediction of the color for each device value to be - displayed - in CIE94 delta E, rather than plain L*a*b* delta E. CIE94 delta E - has a - closer - correspondence with perceived color differences than the default - CIE76 - delta E values.<br> + and the profile prediction of the color for each device value to be + displayed in CIE94 delta E, rather than plain L*a*b* delta E. CIE94 + delta E has a closer correspondence with perceived color differences + than the default CIE76 delta E values.<br> <br> The <b>-k</b> option causes the differences between the test values - and - the profile prediction of the color for each device value to be - displayed - in CIEDE2000 delta E, rather than plain L*a*b* delta E. CIEDE2000 - delta - E has a - closer - correspondence with perceived color differences than either CIE76 or - CIE94 delta E values.<br> - <br> - The <b>-w</b> creates a <a href="File_Formats.html#VRML">VRML</a> - 3D - visualization - of the differences between the test points and the profiles - prediction - of - the resulting colors.<br> - <br> - The <b>-x</b> flag adds Lab axes to the VRML output.<br> + and the profile prediction of the color for each device value to be + displayed in CIEDE2000 delta E, rather than plain L*a*b* delta E. + CIEDE2000 delta E has a closer correspondence with perceived color + differences than either CIE76 or CIE94 delta E values.<br> + <br> + The <b>-w</b> creates a <a href="File_Formats.html#X3DOM">X3DOM</a> + 3D visualization of the delta E's between the test points and the + profiles prediction of the resulting colors.<br> + <br> + The <b>-x</b> flag adds Lab axes to the X3DOM output.<br> <br> The <b>-m</b> flag makes each error line a minimum of 0.5 delta E long, so that all the points are visible. This makes it easier to - view - the distribution of test points in the reference set.<br> + view the distribution of test points in the reference set.<br> <br> The <span style="font-weight: bold;">-e</span> flag causes the - error - vectors in the VRML output to be color coded according to their - lengths, from longest to shortest: yellow, red, magenta, blue, cyan - and - green.<br> + error vectors in the X3DOM output to be color coded according to + their lengths, from longest to shortest: yellow, red, magenta, blue, + cyan and green.<br> + <br> + The <b>-s</b> flag will cause the -v2 patch by patch delta E output + to be sort from largest to smallest. This is useful in identifying + badly read patches.<br> + <br> + The <b>-h</b> flag will display a histogram plot of the fit delta + E's. The X scale is delta E, the Y scale is %<br> + <br> + The <b>-P N.NN</b> option will create a pruned .ti3 file called + data_pN.NN that contains just the measurement points that have a fit + delta E of less than or equal to the given delta E threshold. This + can be useful if you know that there are faulty or poor accuracy + readings in the data set. Use profcheck -h to examine the fit delta + E histogram to choose a threshold that cuts off the tail, and then + profcheck -P to create the pruned data set. You can check that this + is appropriate if a profcheck -h on the resulting profile no longer + has a long tail. <b>Note</b> that using this procedure will be of + no benefit if the tail is due to an inherently poor fit of the + profile to the data rather than reading innacuracy, even if it makes + the fit appear to be better.<br> + <br> + <b>NOTE</b> that the pruning does not take any special care as to + what test points are pruned - it may prune important points such as + <b>white</b> and <b>black</b> points!<br> <br> The <b>-d</b> parameters allow the specification of a particular - device value, - and the test point by test point output will be sorted by distance - from - the - given device value. This can be useful in determining how well - "supported" - the profile is in a particular area of the colorspace.<br> + device value, and the test point by test point output will be sorted + by distance from the given device value. This can be useful in + determining how well "supported" the profile is in a particular area + of the colorspace.<br> <br> If the <b>-p </b>flag is used in combination with the <b>-d</b> - parameters, - then the test point by test point output will be sorted by distance - in - PCS - (Lab) space rather than distance in device space.<br> + parameters, then the test point by test point output will be sorted + by distance in PCS (Lab) space rather than distance in device space.<br> <br> The <b>-f</b> flag enables Fluorescent Whitening Agent (FWA) compensation. This only works if spectral data is available and, the @@ -197,31 +247,27 @@ Sort illumination spectrum to be used as the similated instrument illuminant, overriding the default <b>D50</b> or CIE computation illuminant used for FWA (see <b>-i</b> below<b>). </b>See <a - href="colprof.html#f">colprof -f</a> for - a fuller explanation. The same value should be used as was used - during the creation of the + href="colprof.html#f">colprof -f</a> for a fuller explanation. The + same value should be used as was used during the creation of the profile.<br> <br> The <b>-i</b> flag allows specifying a standard or custom - illumination - spectrum, applied to the spectral test point values to compute CIE - tristimulus values. <b>A</b>, <b>D50</b>, <b>D50M2, D65</b>, <b>F5</b>, - <b>F8</b>, <b>F10</b> are a selection of standard illuminant - spectrums, with <b>D50</b> being the default. If a filename is - specified instead, it will be assumed to be an Argyll specific <a - href="File_Formats.html#.sp">.sp</a> - spectrum file. If FWA compensation is used during measurement, this + illumination spectrum, applied to the spectral test point values to + compute CIE tristimulus values. <b>A</b>, <b>D50</b>, <b>D50M2, + D65</b>, <b>F5</b>, <b>F8</b>, <b>F10</b> are a selection of + standard illuminant spectrums, with <b>D50</b> being the default. + If a filename is specified instead, it will be assumed to be an + Argyll specific <a href="File_Formats.html#.sp">.sp</a> spectrum + file. If FWA compensation is used during measurement, this illuminant will be used by default as the simulated instrument illuminant. The same value should be used as was used during the - creation of the - profile.<br> + creation of the profile.<br> <br> <a name="o"></a> The <b>-o</b> flag allows specifying a tristimulus observer, and is used to compute PCS (Profile Connection Space) tristimulus values. The following choices are available:<br> <b> 1931_2</b> selects the standard CIE 1931 2 degree - observer. - The default.<br> + observer. The default.<br> <b>1964_10</b> selects the standard CIE 1964 10 degree observer.<br> <b>1955_2</b> selects the Stiles and Birch 1955 2 degree @@ -232,16 +278,15 @@ Sort observer<br> <br> The same parameter value should be used as was used during the - creation - of the profile.<br> + creation of the profile.<br> <br> The <span style="font-weight: bold;">-I</span> parameter allows changing the intent used in looking up the ICC profile colors to relative colorimetric. This would <span style="text-decoration: - underline;">not</span> be used if you are - checking a profile against the .ti3 file that was used to create it, - since, since profiles are always made - from absolute colorimetric measurement values.<br> + underline;">not</span> be used if you are checking a profile + against the .ti3 file that was used to create it, since, since + profiles are always made from absolute colorimetric measurement + values.<br> <br> <br> <br> |