The key advantage of logarithmic wedges is that charts with large f_max/f_min ratios work well for systems with a wide range of resolutions, unlike hyperbolic wedges, where large ratios cause high frequencies occupy excessive real estate and low frequencies to become highly compressed. The f_max/f_min ratio of the hyperbolic wedges in the ISO 12233/2017 chart is only 12.5:1 — insufficient for modern high resolution cameras.

With logarithmic wedges, ratios of 50:1 or greater are practical because frequency bands characterized by an f_max/f_min ratio (for example, 10:1 for decades or 2:1 for octaves) occupy the same amount of space, as they do in most frequency response curves and Bode plots. Frequency bands occupy neither too much nor too little space.

We begin by illustrating the hyperbolic wedge, using a portion of the eSFR ISO chart.

Hyperbolic wedges

The first image shows several wedges in the standard eSFR ISO chart, which have a maximum frequency of 2500 Line Widths/Picture Height (LW/PH) when the chart is framed so the inside black borders are located at the top and bottom of the image. This scaling, indicated by the label “25”, has been used since ISO 12233:2000. Because 2500 LW/PH is inadequate for modern high resolution cameras, Imatest also offers a 4000 LW/PH chart. The frequency ratio for the high frequency wedges is 12.5:1. In practice it’s not necessary to frame the eSFR ISO chart in this way because Imatest derives wedge frequencies directly from the image.

Crop of eSFR ISO chart with hyperbolic wedge maximum frequency 2500 LW/PH

The CIPA Visual Resolution test chart from ISO 12233:2017 §5 and Annexes A and B is shown on the right, along with a slice of an eSFR ISO chart containing similar wedges. Both these charts have frequency labels in the larger wedges from 2 to 25 (a 12.5:1 ratio), representing 200 to 2500 LW/PH when the chart is framed so the inside black borders are located at the top and bottom of the image, as indicated above. For this framing, the wedge will not reach the Nyquist frequency (fNyq = 0.5 C/P) for image sensor heights larger than 2500 pixels (about 9.4MP for cameras with a 3:2 aspect ratio). Note that the height, not the width, of the wedge is significant. The ISO wedges were designed during digital photography’s distant past, when 10MP consumer cameras were a distant dream. They’re inadequate for recent high resolution cameras.

Because 2500 LW/PH wedges are inadequate for modern high-resolution cameras, we designed a version of the eSFR ISO chart, shown below, with a maximum frequency of 4000 LW/PH. 

Crop of eSFR ISO chart with hyperbolic wedge maximum frequency 4000 LW/PH

This chart had a problem with a relatively low resolution (720P) camera.  Because the low frequencies are highly compressed and the image was blurry, there were not enough rows of distinct bars to reliably calculate the frequencies. Spatial frequencies between 2000 and 4000 LW/PH take up over half the wedge; much more than needed.

With hyperbolic wedges we need more than one pattern to accommodate the wide range of camera resolutions we typically see. A “one size fits all” pattern would be highly desirable. To get it we developed an eSFR ISO chart with…

Logarithmic wedges

For logarithmic wedges, \(f = f_0 10^{k(x-x_0)}\), where f is spatial frequency, and \(f = f_0 \text{ at } x = x_0\) (the lowest spatial frequency).

For frequency f,  \(x = x_0 + \log_{10}(f / f_0) / k\). A ratio of any two frequencies f₂/f₁ will occupy a constant distance \(10^{k(x_2-x_1)}\).

Crop of eSFR ISO chart with logarithmic wedge maximum frequency 5000 LW/PH

Because this chart has an improved distribution of frequencies over the available space, with octaves (f_max/f_min ratios of 2:1) or decades (f_max/f_min ratios of 10:1) occupying the same amount of space, it avoids the two problems we experienced with hyperbolic wedges. 

• The maximum frequency can be made very high without hogging excessive real estate. This makes it easy to design wedges for ultra-high resolution systems.
• Low frequencies increase slowly enough to comfortably accommodate low resolution images.

Trapezoidal edges, which consist of converging straight lines, are rarely used in modern charts because they have relatively little real estate for the important high frequencies. If you’re old enough, you might remember seeing them in TV test patterns. Since they’re rarely used in modern test charts, we won’t discuss them further.

Results

Starting with 2021.2, Imatest automatically detects the wedge type— hyperbolic or logarithmic. We present results for synthesized images of the 4000 LW/PH hyperbolic wedge and the 5000 LW/PH logarithmic wedge. 3200×1800 pixel Extended eSFR ISO charts with hyperbolic and logarithmic wedges were created with Imatest Test Charts, converted to bitmaps in Inkscape, then blurred (Gaussian filter 1 = 0.7 sigma) with a little noise added (0.0005 sigma, 0.5 luma fraction) in Imatest Image Processing.

Note that for the 1800 pixel high image, Nyquist frequency (always 0.5 Cycles/Pixel) = 1800 Line Widths per Picture Height. The difference in appearance for the two wedge types is more prominent in high frequency the wedges on the right, below. (The difference is less obvious for the low frequency wedges, which have a frequency ratio of only 4:1).

Results for the hyperbolic wedge

Results for the logarithmic wedge

The results are very close: nearly identical except for some details (like a strange boost around 0.55 to 0.6 C/P), which is outside the range of important measurements. The main causes of the differences are noise and phase of the bars. Wedges tend to produce less consistent MTF measurements than slanted-edges.

Summary

Logarithmic wedges have two  closely-related advantages: (1) they allow the design of high ratio f_max/f_min charts where low frequencies aren’t overly compressed and high frequencies don’t occupy excessive area, and (2) they accommodate a wide range of image resolutions from very low to very high. This is the result of their improved distribution of frequencies— similar to standard frequency response plots.

Here is an example from the original version of the 12233:2000 chart (designed back in the day when 2MP was “high” resolution). It has two wedges covering frequencies from 100-2000 LW/PH — inadequate for modern high resolution cameras. The two wedges could be replaced by a single wedge with frequencies from 100-5000 (or greater) LW/PH, shown on the right. 

Segment of ISO 12233:2000 chart showing two hyperbolic wedges, 100-600 and 500-2000 LW/PH

Logarithmic wedge, 100-5000 LW/PH

The post Logarithmic wedges: a superior design appeared first on Imatest.

Recent speed improvements allow for real-time focusing and allow users to analyze images from two types of sources:

• Direct image acquisition from a variety of devices, listed in detail in Supported image acquisition hardware for Imatest Master. 
• Image files listed in Image file formats and acquisition devices,

Although the majority of images traditionally analyzed by Imatest have been from files (JPG, PNG, etc.), three modules, which can perform a majority of Imatest’s analyses, support direct data acquisition, and can be used for realtime analysis.

• Rescharts, which analyzes a variety of sharpness/MTF charts: slanted-edge charts with automatic or manual detection, Siemens Star, Log Frequency Contrast, Spilled Coins (Dead Leaves), or any arbitrary image (which produces uncalibrated relative sharpness measurements). Depending on the chart, measurements may include lateral chromatic aberration, optical distortion, and tonal response (OECF). Noise, SNR (Signal-to-Noise Ratio), and color response are available with eSFR ISO.
• Color/Tone Interactive, which analyzes all widely-used color and grayscale charts (as well as a good many custom or obscure charts). Measurements include tonal response, color accuracy, noise, and SNR,
• Uniformity Interactive, which analyzes flat field images for vignetting (light falloff), color nonuniformity, noise, SNR, and PRNU (photo response nonuniformity) and DSNU (dark signal nonuniformity).

While all measurements have been sped up by a factor of 4 to 10 or more compared with 2020.1, focusing is the most important and most speed critical function. If it’s done at all, it needs to be fast.

The following video, made using an inexpensive 2MP USB camera selected with the Device Manager and run in Rescharts, illustrates real-time focusing using slanted edges from an eSFR ISO test chart.

Video: Realtime focusing with direct data acquisition

Direct image acquisition works as shown in the illustration below. 1. Acquire the image. 2. Analyze. 3 Display. Then repeat until Stop reload is pressed.

Here are the key settings for direct image acquisition with real-time updates.

1. Run Device Manager to select the device to analyze, the device resolution, and to prepare the image using the Preview function.
2. Open one of the three modules that support image acquisition: Rescharts, Color/Tone Interactive, or Uniformity Interactive. Note that SFRplus Setup, eSFR ISO setup, SFRreg Setup, and Checkerboard Setup all run Rescharts.
3. In the Settings dropdown menu, set (check) Auto reload. (The setting will be saved; you won’t need to repeat this.)
4. Acquire the image.
5. Select the display you want to see reloaded, then press Reload.
6. The Reload button will change color to Yellow and indicate the number of reloads followed by “Stop reload“. Press Stop reload when you are finished.

Full instructions are on Using Direct Image Acquisition.

The post Real-time focusing with Imatest Master direct data acquisition appeared first on Imatest.

How different are the reference and camera colors in the Colorchecker image on the right, represented in these diagrams?

Color differences and MacAdam ellipses

When these representations are viewed, the question naturally arises, “how different are the reference and camera colors?” Color differences can be quantified by several measurements— ΔE_ab,  ΔE₉₄,  and ΔE₀₀ (where 00 is short for 2000), where ΔE measurements include chroma (color) and luminance (brightness). If brightness (L*) is omitted, these measurements are called _ΔCab,  ΔC₉₄,  and ΔC_00, where C stands for chroma (which includes a color’s hue and saturation). In this post we will discuss chroma differences. ΔC_ab = (a*²+b*²)^1/2 (sometimes called ΔC) is the simple Euclidean (geometric) distance on the a*b* plane. It’s familiar but not accurate.

Starting with Imatest 4.2, Imatest’s two-dimensional chroma displays— CIELAB a*b*, CIE 1931 xy chromaticity, CIE u’v’ chromaticity, Vectorscope, and CbCr— can display ellipses that assist in visualizing perceptual color differences.

These ellipses were developed from the MacAdam ellipses, shown on the right.

MacAdam ellipses shown in the
CIE 1931chromaticity diagram,
magnified 10X (from Wikipedia)

The MacAdam ellipses were developed from a set of experiments performed at the University of Rochester in 1942, in which an observer tried to match pairs of colors, one fixed and one variable. The ellipse parameters are based on statistical variations in the matching, which are closely related to Just Noticeable Differences (JND). Twenty-five colors (whose xy values are shown as • in the illustration) were used.

Almost all images of MacAdam ellipses show the ellipses for the original 25 colors. In imatest we use a sophisticated interpolation routine to determine the ellipse parameters for color test charts. This is quite reliable since the gamut of the original color set extends well beyond the gamut of the widely-used sRGB color space (the standard of Windows and the internet), as well as most other color spaces used in imaging. At least 10 of the original colors are outside the sRGB gamut. sRGB is used for all examples in this post.

The ellipses are a visual indicator of the magnitude of perceived color (chroma) difference. The longer the ellipse axis, the greater the distance on the a*b* plane for a given color difference.

Here are the MacAdam ellipses for the X-Rite Colorchecker, in xy (from xyY) and u’v’ (from Lu’v’) representations, displayed in Multicharts. u’v’ is supposed to be more perceptually uniform than xy, and that is not evident inside the sRGB gamut. This may be one reason why u’v’ hasn’t gained much traction in the imaging industry.

xy Colorchecker MacAdam ellipses, magnified 10X

xy Colorchecker MacAdam ellipses, magnified 10X

The MacAdam ellipses are not widely used in imaging (though they seem to have traction in LED lighting). Instead, color differences based on CIELAB (L*a*b*) color space, which was developed in 1976 with the intent of being much more perceptually uniform than xyY, are used. Color differences are usually presented as ΔE_ab,  ΔE₉₄,  or ΔE₀₀. In all ΔE (color difference) equations, the Luminance (L*) term can be easily removed, leaving chroma differences ΔC, which can be displayed in in two dimensions.

Color difference ellipses in Imatest

All Imatest modules that can produce two-dimensional color different plots can display color difference ellipses. Colorcheck, SFRplus, and eSFR ISO display them in the a*b* plot. Multicharts and Multitest display them in a*b*, xy, u’v’, Vectorscope, and CbCr plots. In addition to MacAdam ellipses, which are not generally used for color difference measurements (but are interesting for comparison), the following color difference metrics are presented. The illustrations below are for a Colorchecker (shown above) analyzed in Multicharts.

ΔC_ab (plain ΔC – 1976)

ΔC_ab is simple geometric distance in the a*b* plane of CIELAB (L*a*b*) color space. When CIELAB was designed, the intent was that ΔC = 1 would correspond to one Just Noticeable Difference (JND). This may hold for colors with very low chroma (a*²+b*²)^1/2, but it fails badly as chroma increases, which is why ΔE₉₄  and ΔE₀₀ were developed. ΔE_ab and ΔC_ab are familiar and often mentioned in imaging literature, but they should never be used for important measurements.

a*b* Colorchecker ΔCab ellipses (circles!), magnified 4X

xy Colorchecker ΔCab ellipses, magnified 4X, zoomed in

ΔC ellipses are always show magnified 4X (10X would be too large for clear display). We show the xy plot zoomed in so more detail is visible.

ΔC₉₄ (ΔC 1994)

ΔE₉₄ was developed to compensate for the deficiencies of ΔE_ab. Circles become ellipses with their major axes aligned with the radius from the origin (a* = b* = 0). It is more more accurate than ΔE_ab, especially for strongly chromatic colors.

a*b* Colorchecker ΔC94 ellipses, magnified 4X

xy Colorchecker ΔC94 ellipses, magnified 4X, zoomed in

ΔC₀₀ (ΔC 2000)

ΔE₀₀ was developed as a further refinement over ΔE₉₄. It has an extremely complex equation. Though not perfect, it’s the best of the current color difference formulas, and is recommend when color differences need to be quantified. It’s quite close to ΔE₉₄, except for some blue colors. Until we added these ellipses to Imatest, there was no convenient way to visually compare the different ΔE measurements!

a*b* Colorchecker ΔC00 ellipses, magnified 4X

xy Colorchecker ΔC00 ellipses, magnified 4X, zoomed in

Displaying the ellipses in Imatest

In Imatest 4.2 the ellipses are not displayed by default. We may change this in a later version. The table below shows how to control ellipse display.

Multicharts Multitest	Multicharts and Multitest share ini file settings. Ellipse settings for both are set in the Color ellipses dropdown menu in Multicharts. Standard ellipse magnification (10x for MacAdam ellipses, 4X for Delta-C ellipses) is recommended.
Colorcheck	The selection is in the middle-left of the Colorcheck settings window.
SFRplus eSFR ISO	In (Rescharts) SFRplus and eSFR ISO, ellipse settings are located in the Ellipses dropdown menu, displayed to the right of the bottom of the a*b* plot, when it is selected.

The post Color difference ellipses appeared first on Imatest.

Test images at infinity distance (or any distance of choice) using a compact projection system such as the Optikos Meridian camera test system. Test the sharpness of extreme fisheye lenses (with angular fields of view over 180 degrees, whose MTF cannot be measured near the image boundaries with a single flat target) using an array test charts, each consisting of an individual registration mark. Registration mark charts, such as the one shown on the right, may be purchased from the Imatest store or printed on a high-quality inkjet printer. Since Region of Interest (ROI) selection is automatic, they may be positioned where needed. They work best when facing the camera.

Registration mark

Details of the regions to be analyzed are based on user-entered criteria (similar to SFRplus or eSFR ISO, which it closely resembles).

Sharpness is derived from light/dark slanted edges inside the registration marks, as described in Sharpness: What is it and how is it measured? SFRreg can handle a wide range of camera aspect ratios and chart arrangements.

SFRreg operates in two modes.

• Interactive/setup mode  allows you to select settings and interactively examine results in detail. Saved settings are used for Auto Mode.
• Auto mode  runs automatically with no additional user input. ROIs are located automatically based on settings saved from the interactive/setup mode. This allows images of different sizes and framing to be analyzed with no change of settings. Auto mode works with large batches of files, and is especially useful for automated testing, where framing may vary from image to image.

Part 1 of the instructions introduces SFRreg and explains how to obtain and photograph the chart. Part 2 shows how to run SFRreg inside Rescharts and how to save settings for automated runs. Part 3 illustrates the results.

SFRreg images from the Optikos Meridian projection system Imatest SFRreg is originally designed designed to work with images from the Optikos Meridian system, which consists of several projectors that project registration mark patterns towards a camera. These patterns appear at infinity focus at the camera. A typical image is shown below.

Image acquired from a 9-projector Optikos Meridian system

SFRreg images from arbitrary arrays of printed registration mark charts

SFRreg also works with printed registration mark patterns, which can be placed anywhere in the image. For extreme wide-angle (fisheye) lenses they should be oriented facing directly towards the camera. Here is a synthesized image (we’ll add a real one soon). You can add other charts— typically color or grayscale— to the image for additional measurements.

Fisheye lens image with synthesized registration mark charts oriented facing the camera

SFRreg chart print options (can be selected when ordering)

	Options	Notes
Media	Inkjet (reflective), LVT film (transmissive)	Inkjet (reflective) is usually the most practical choice.
Contrast	4:1, 10:1	4:1 contrast is specified in the new ISO 12233:2014 standard.
Surface	Matte or semigloss	Semigloss is slightly sharper, but is more susceptible to glare (specular reflections), especially with wide angle lenses. Matte surface is recommended for wide angle lenses or difficult lighting situations.

 

Slanted-edge algorithm The algorithms for calculating MTF/SFR were adapted from a Matlab program, sfrmat, written by Peter Burns () to implement the ISO 12233:2000 standard. Imatest SFR, SFRplus, SFRreg, and eSFR ISO incorporates numerous improvements, including improved edge detection, better handling of lens distortion, a nicer interface, and far more detailed output. The original Matlab code is available on http://losburns.com/imaging/software/SFRedge/index.htm. In comparing sfrmat results with Imatest, note that if no OECF (tonal response curve) file is entered into sfrmat, no tonal response curve is assumed, i.e., gamma = 1 (linear response). Since the default value of gamma in Imatest is 0.5, which is typical of digital cameras, you must set gamma to 1 to obtain good agreement with sfrmat.

Multi-ROI summary display

 

SFRreg results in Rescharts window: Multiple region (ROI) summary
(Only upper Vertical regions have been selected to keep the view uncluttered.)

The multi-ROI (multiple Region of Interest) summary shown in the Rescharts window (above) contains a detailed summary of SFRreg results. (3D plots also contain an excellent summary.) The upper left contains the image in muted gray tones, with the selected regions surrounded by red rectangles and displayed with full contrast. Up to four results boxes are displayed next to each region. The results are selected the the Display options area on the right of the window, below the Display selection.

The Results selection (right) lets you choose which results to display. N is region number. Ctr-corner distance % is the approximate location of the region. CA is Chromatic Aberration in area, as percentage of the center-to-corner distance (a perceptual measurement). A legend below the image shows which results are displayed. The View selection (far right) lets you select how many results boxes to display, which can be helpful when many regions overlap. From top to bottom the number of boxes is 4, 3, 2, 2, and 1, respectively.

Results selection

View selection

The post SFRreg: SFR from Registration Marks appeared first on Imatest.

Here is a crop of a multiburst pattern, generated by a Tektronix device:

Crop of Multiburst pattern: click on image for full-size pattern that you can download.
Running Stepchart with the multiburst pattern

Open Stepchart, then read the multiburst image. Make a rough region selection for the multiburst pattern— it will be refined as shown below. Then press Yes (not Express mode) to bring up the Stepchart settings window.

Stepchart settings for Multiburst MTF measurement

The key settings are circled in red. Automatic (Zone detection) should be unchecked and 6 patches should be selected using the slider (for this pattern; some multiburst patterns have as few as 5 zones). Under Results (Fig. 1, Lower plot), select 4. Noise uncorrected for patch nonuniformity, norml. to 1 in patch 1. Only Plot 1. Pixels, noise is relevant to the multiburst image. When settings are complete, click OK. If this is the first run for this size of image, the fine ROI adjustment window will appear.

Fine ROI adjustment window for Multiburst pattern in Stepchart. May be enlarged.

Results

The standard deviation of the pixels in the patch, which is normally used to measure noise, is proportional to the MTF of the pattern in the patch (as long as it’s above the noise). To obtain an accurate result the patch nonuniformity correction is turned off (by setting Results (Fig. 1, Lower plot) to 4. Noise uncorrected…, as indicated above.

The key results are in the first Stepchart figure, most importantly in the lower plot, which contains the MTF for the patches.

Stepchart Multiburst results: MTF is in lower plot

The post Measuring Multiburst pattern MTF with Stepchart appeared first on Imatest.

Equations – Applying the correction factor – Verification 

The LSF correction factor primarily affects very high spatial frequencies beyond most of the energy for typical high quality cameras. But it does make a difference for practical measurements: MTF50 for a typical high quality camera (shown below) is increased by about 1.5%.

The correction factor is turned on by default in Imatest 4.2+. It is set by pressing Settings, Options III in the Imatest main window as shown below, then checking the box for the correction. We strongly recommend leaving on the LSF correction factor, i.e., the box should be checked.

 
More details about Imatest’s ISO compliance can be found on the Sharpness web page.

Equations

Show the SFR and LSF equations from ISO 12233:2000, 2014, and 2017Hide the SFR and LSF equations from ISO 12233:2000, 2014, and 2017

The difference between the 2014 and 2017 equations is the LSF (Line Spread Function) correction factor D(j) (D(k) in ISO 12233:2017), which corrects for the numerical differentiation used to calculate LSF (not the point spread function) from the Edge Spread Function. Note that D(j) (or D(k) ) are incorrect in both the 2014 and 2017 standards. They should read,

Apart from min[…,10], this equation is used in ISO 12233:2022.

Applying the correction factor

The LSF correction factor is enabled by default. To check or change the setting, click Settings (in the Imatest main window), Options III. You can check or uncheck the Use LSF… checkbox as appropriate, but we recommend leaving it checked. The checkbox sets derivCorr in the [imatest] section of imatest-v2.ini to 0 (correction off) or 1 (correction on).

Options III window for applying or removing LSF correction factor

Verification

To observe the effects of the LSF correction factor, we use an idealized edge, tilted 5 degrees, shown on the right. You can click on it to download it for your own testing.

The ideal edge increases uniformly from 0 to 1 over a distance of τ = 1 pixel.

Here are the results without and with the LSF correction factor. Note that gamma has been set to 1 because the idealized image is not gamma-encoded. The results with LSF correction are much closer to the expected value of 2/π = 0.6366. The difference is likely due to digital sampling and the numerical binning/oversampling process: the average oversampled edge shown in the figures below is slightly rounded. Also, the edge rotation correction that is not applied to edges slanted by less than 8 degrees. The edge rotation is not included in the ISO standard and so is not applied to edges that fall under the ISO standard algorithm.

Ideal edge, uncorrected MTF. MTF@Nyquist = 0.56.

Ideal edge, LSF-corrected MTF. MTF@Nyquist = 0.624.

The figure on the right, generated by MTF Compare (a postprocessor to MTF calculation programs for comparing MTF calculations), compares the uncorrected MTF of the ideal edge (blue) with the corrected MTF (burgundy). The black line is the uncorrected/corrected transfer function = 1/(correction factor D). It has the expected values of 0.9 (1/1.1107) at the Nyquist frequency (f = 0.5 C/P) and 2/π = 0.6366 at 2*Nyquist (1 C/P).

MTF Compare Uncorrected vs. Corrected

Effects of gamma Most color space files are gamma-encoded with gamma around 0.5, which is the default for Imatest slanted-edge SFR calculations. (Actual gamma varies considerably, and may be complicated by a tonal response curve on top of the gamma curve.)  However, the ideal pulse is not gamma-encoded, so gamma = 1 is the appropriate setting. If gamma is set to 0.5 (the default), an inappropriate linearization will be applied to the edge, resulting in erroneous results: the MTF at 2*Nyquist is not equal to 0, as it should be.

Ideal edge, uncorrected MTF, calculated (incorrectly) with gamma = 0.5

The idealized image has much more energy above the Nyquist frequency than typical high quality images. Here is an example for a high quality camera showing the effects of the LSF correction factor on MTF50— the most commonly-used summary metric.

Typical edge, uncorrected MTF. MTF50 = 0.355 Cycles/Pixel.

Typical edge, LSF-corrected MTF. MTF50 = 0.360 C/P:  1.41% higher.

 The difference may not be significant for many applications.

Consistency with older Imatest versions

The new correction factor is turned on by default in Imatest 4.2+ releases.

You can override the default, i.e., turn the correction factor on or off, by clicking Settings (in the Imatest main window), Options III, and checking or unchecking the Use LSF… checkbox as appropriate. Once the setting is saved, it will be retained across all future versions unless changed by the user. The default calculation be selected unless you manually change it in Options III window.

We strongly recommend leaving on the LSF correction factor, i.e., the box should be checked.

For industrial testing with pass/fail thresholds set up with the old calculation, we recommend either:

• adjusting your pass/fail specification to account for the measurement change
• continuing to use the old calculation by setting devCorr = 0 in the [imatest] section of the ini file

For example, on an ideal edge MTF at Nyquist/4 (0.125 C/P, a common KPI) is only increased by 0.64%,  at Nyquist/2 (0.250 C/P) MTF is increased by 2.80%. On an actual sharp, RAW camera-phone image with Gamma=1, the MTF at Nyquist/4 is increased by 0.81%, MTF at Nyquist/4 is increased by 3.00%.

The post LSF correction factor for slanted-edge MTF measurements appeared first on Imatest.

*F is an abbreviation for Frequency in Log F-Contrast.

I was motivated to write this post by my curiosity about the 36-megapixel mirrorless full-frame interchangeable-lens Sony Alpha A7R  and how its image quality compares to competitive cameras (as well as cameras I now own). A post on Imaging-Resource.com says,

“In our initial press briefing, Sony made quite a big deal about its new sharpening algorithms, designed to reduce what they call “outlines” — which we’ve called “halos” for some years now. The images shown in that briefing were certainly impressive, but it’s hard to judge when there’s no standard of comparison against other cameras shooting the same scenes under controlled conditions. We had to see how it did on our own standardized studio shots before we could pass final judgment.

We’ve now done just that, and have to say we’re pretty well blown away by what we saw. Check our Sony A7R review page for our initial image comparison analysis…”

I will compare objective measurements from Log F-Contrast chart images. These provide a useful quantitative supplement to Imaging-Resource’s subjective (but vital) camera comparisons that use still-life images, enabling us to accurately asses Sony’s claim of sharpening without “halos”, to examine how noise reduction affects fine detail, and to compare different types of cameras. I’ll report on cameras of particular interest to me as well as cameras that I own.

My quest may be familiar to many of you: a small lightweight camera that can  produce images comparable to the large-format prints by Ansel Adams and Edward Weston that I saw at the George Eastman House when I was growing up in Rochester, NY. In a word, I’m looking for the holy grail— at an affordable price. A modest quest.

Imaging-Resource.com

Imaging-resource.com publishes outstanding camera reviews, indexed on their Camera Reviews page. To find the Log F-Contrast images, go to a review page (here for the Sony A7R), click on the Samples tab, then scroll down (about 3/4 of the way to the bottom for the A7R). Look for the Multi Target image. Click on a thumbnail taken at a low ISO speed to see the camera’s ultimate capability. Later you may want to load images taken at higher ISO speeds to see how signal processing changes the results (you’ll almost always see additional noise reduction that tends to reduce sharpness, especially in low-contrast regions). Here is an image Imaging Resource’s Multi Target, showing the Log F-Contrast chart just below the center.

Imaging-Resource Multi Target image

The Log F-Contrast chart

The Log F-Contrast chart increases in spatial frequency from left to right and decreases in contrast from top to bottom. (Actually the square of the contrast is proportional to the distance from the bottom, so contrast in the middle is 1/4 of the contrast at the top. This makes the chart look much better.)

Log F-Contrast chart image

Details of running the Log F-Contrast are given in Log F-Contrast instructions. We focus on results here. The image below shows the normalized MTF* for several rows in the chart from near the top (row 1 of 25) to about 80% of the way to the bottom (row 19 of 25). (MTF curves for rows below this are unreliable due to noise and low contrast.)

*MTF (Modulation Transfer Function), which is generally equal to SFR (Spatial Frequency Response)
is a measure of image sharpness, described in Sharpness: What is it and how is it measured?

Sony Alpha A7R

Sony A7R:  Normalized MTF for bands from just below the top to near the bottom

MTF (i.e., sharpness) is fairly consistent for rows 1 through 10 (the relatively high contrast portion near the top of the image), with no significant sharpening peaks (Sony’s claim of no “halos” is well-justified). The long MTF ramp is an indication of nicely done sharpening. Sharpness starts to fall at row 13 (1.7:1 contrast), and drops significantly for rows 16 and 19.

This is the result of noise reduction, which is equivalent to lowpass filtering (smoothing), and which depends on image content in most consumer cameras. Noise reduction is minimum (and its inverse, sharpening, i.e., high frequency boost, is maximum) near contrasty features— near the top of the chart. Noise reduction is strongest in smooth areas— in the absence of contrasty features— near the bottom of the chart.

Comparison with slanted-edge measurements Results from Log F-Contrast illustrate how sharpening increases with feature contrast in the JPEG output files of most consumer cameras. Slanted-edges tend to maximize sharpening, especially if they are contrasty. Imaging-Resource has an A7R image of the ISO-12233 chart, which has several high-contrast edges, making for a convenient comparison. The average edge and MTF are shown on the right. Results are quite close to row 1 of the Log F-Contrast results (blue curve, above) up to about 3000 LW/PH, but the slanted edge has higher MTF at higher spatial frequencies. It has quite a lot of energy above Nyquist, which is characteristic of a system with no anti-aliasing filter + a very sharp lens. This can be an indicator of aliasing (color moire), but none is visible in the image. This is very impressive performance. For raw images MTF tends to be similar for different charts because little or no sharpening or noise reduction is applied. Different techniques and charts for measuring sharpness— and how results are affected by image content— are compared in the MTF measurement matrix.

Since the multiple MTF image (above) is somewhat difficult to interpret, Imatest produces a contour plot of normalized contrast (MTF) over the image. This plot clearly shows how MTF varies with chart contrast.

 Sony A7R:  Normalized contrast plot

This display is clearer than the MTF plot above: it contains results for all rows with sufficient contrast for MTF to be measured. It’s much easier to see how MTF drops as contrast decreases. It also displays key summary metrics: MTF50 (the spatial frequency where contrast falls to 50% of its low frequency value), MTF50P (the spatial frequency where contrast falls to 50% of its peak value), and comparable numbers for 20 and 10% contrast (which are more closely related to “vanishing resolution”). MTF50P is the most robust summary metric than MTF50: it is less sensitive than MTF50 to strong sharpening (not present here, but significant in the Canon EOS-6D).

Significant texture loss due to noise reduction is visible for modulation under 0.2 (chart contrast ratio under 1.7:1). This typical behavior for JPEG processing. If you want this detail you’ll have to convert raw images off the camera (with noise reduction turned down). The MTF50P values over 3,000 Line Widths/Picture Height are outstanding, particularly considering that there is no sharpening peak. With good technique you can expect razor-sharp 24×32 inch prints (the largest you can print in this format on a 24-inch printer)— and you can go a lot larger with excellent quality. How many photographers actually make larger prints?

Regarding Sony’s claim of an excellent sharpening algorithm with no (significant) halo, we know from slanted-edge analysis that MTF (spatial frequency response) peaks above 1 correspond to halos near edges, and that the strength of the halo is well correlated with the strength of the peak. Halos corresponding to MTF peaks under 1.1 (10% above the low frequency MTF) are generally too weak to be objectionable. So the verdict on the Sony A7R is yes, it lives up to its claims, but it’s hardly unique in that regard: the Pentax 645D, the Nikon D800 and some others (but notably not the Canon EOS-6D), also have excellent sharpening without a significant sharpening peak.

Pentax 645D

The 40-megapixel medium-format Pentax 645D is the camera to beat.

Pentax 645D:  Normalized contrast plot
(Note that the horizontal scales are different for each plot: the maximum frequency is the Nyquist frequency.)

At high contrast levels the Sony does indeed beat the Pentax! But the Pentax appears to have NO noise reduction: MTF doesn’t drop until modulation gets down to 0.03— an extremely low level. The Pentax wins for reproducing extremely fine texture at modulation levels below 0.2. I would guess that a properly processed raw image from the Sony A7R would have comparable low-level detail (see the RX100 II raw image, below), though there might be more noise (which might not be objectionable at this ISO speed).

The Sony holds its own very well against the Pentax, which is particularly impressive considering its size, weight, and cost. The difference would only be noticeable in very large prints.

 Nikon D800

Unfortunately I couldn’t find a Multi Target image for the D800E (the version without the anti-aliasing (optical lowpass) filter). This would have made for a much better comparison.

Nikon D800:  Normalized contrast plot
(Note that the horizontal scales are different for most plots, but this is the same as the Sony A7R.)

Signal processing is quite similar to the Sony A7R, but MTF frequencies are about 15% lower, which is what you’d expect from the added anti-aliasing (optical lowpass) filter. We would expect performance comparable to the A7R with the D800E, which doesn’t have an anti-aliasing filter. Of course the lens could have some effect, though excellent lenses at or near optimum aperture have been used in both cases (and in other Imaging-Resource images). The lens is probably not the dominant factor here.

Canon EOS-6D

Canon’s JPEG processing— and I mean sharpening, not JPEG compression— masks the 6D’s full performance potential. (It could probably be corrected in firmware if Canon wanted.) To begin with, the image is strongly oversharpened. Halos will be clearly visible in moderately large prints (as they are in IR’s image). And the sharpening radius is 2, which boosts middle but not the highest spatial frequencies (explained here). A sharpening radius of 2 is good for moderate enlargements with average quality lenses, but it’s hardly optimum when maximum quality is required. I never use JPEGs for high quality prints from the 6D; I always convert from raw and sharpen with a smaller radius with excellent results— I’ve made some outstanding 16×24 inch prints from the 6D.

Canon EOS-6D:  Normalized contrast plot
(Note that the horizontal scales are different for each plot: the maximum frequency is the Nyquist frequency.)

Despite the sharpening, MTF50P is well below that of the Sony A3R. This is because of the way Canon’s sharpening (with radius = 2) works, boosting middle but not high spatial frequencies, and partly because the Sony has more pixels combined with an excellent lens and no anti-aliasing filter. With raw images the difference would be reduced, but the Sony is the clear winner here.

Panasonic Lumix DMC-G3

The G3 is a 16-megapixel Micro Four-Thirds mirrorless interchangeable system camera that I’ve had since 2011 (two years ago as I write this). It’s much smaller and lighter than than APS-C or full-frame DSLRs; it’s easier to carry on long trips.

Panasonic Lumix DMC-G3:  Normalized contrast plot
(Note that the horizontal scales are different for each plot: the maximum frequency is the Nyquist frequency.)

MTF50P in the range of 2200 is respectable for a camera in this category, though not in the same class as the Sony, Pentax, and Nikon cameras above. There is a moderate sharpening peak— not enough to be objectionable in any viewing conditions. Noise reduction is strong for modulation under 0.2— pretty similar for most of the cameras in this post, except for the Pentax. I’ve made some very nice 18×24 inch prints with this camera, using an off-camera raw converter (though they’re not quite as good as processed raw images from the EOS-6D— not JPEGs).

Panasonic Lumix DMC-LX7

Nobody, and I mean nobody, will take you for a serious photographer if you carry one of these little toy-like cameras. But appearances can be deceiving. A toy it is not. This 10-megapixel point and shoot (with an f/1.4 Leica zoom lens) is capable of some pretty impressive image quality for a slightly larger than pocketable point-and-shoot. I carry this camera when I’m feeling lazy but want to make fine images should the opportunity arise.

Panasonic Lumix DMC-LX7:  Normalized contrast plot
(Note that the horizontal scales are different for each plot: the maximum frequency is the Nyquist frequency.)

Sharpening is rather strong, with the sharpening peak reaching 1.37, which might be objectionable in large prints. (Raw is available.) It might be a good idea to decrease sharpening from the default level. MTF50P above 2000 at high contrast levels is excellent for a 10 megapixel camera, thanks to an excellent lens and strong sharpening with radius R = 1 (which does a much better job than the R = 2 sharpening in the EOS-6D, as explained in the Sharpening page).

Sony RX100 II

My London son is absolutely delighted with his 20-megapixel RX100 (with a Zeiss f/1.8 zoom lens), which he first used to take pictures of the aurora borealis in Iceland (an application that requires excellent low light performance). It’s small enough to fit in most pockets (slightly smaller than the LX7). We’ll look at the RX100 II, which is the current model. It has a large (1-inch) sensor, unusual for such a small camera.

Sony RX100 II:  Normalized contrast plot: JPEG image
(Note that the horizontal scales are different for each plot: the maximum frequency is the Nyquist frequency.)

Note that MTF50P for modulation above 0.1 is better than the Canon EOS-6D, despite the Canon’s strong sharpening. Impressive for a camera you can put in a pocket! Only at very low contrast does the big Canon out-perform the little Sony. The Sony is also quite a bit better (higher MTF50P with less sharpening) than the slightly-larger but less expensive Panasonic LX7.

The reduced MTF at very high modulation levels (above 0.8) is caused by signal processing— I’m not sure why. Imaging-Resource has supplied a raw (ARW) version of this image, which we converted using dcraw, which has no sharpening or noise reduction. Here is the plot, which shows a MTF falloff characteristic of an unsharpened image and vertical contour lines characteristic of no noise reduction

Sony RX100 II:  Normalized contrast plot: raw (ARW) image converted by dcraw

The Sony RX100 II hits a real sweet spot: it’s small enough to fit in most pockets and capable of making serious enlargements. It may well meet the holy grail requirements. If nothing else, the new crop of cameras removes technical excuses for not making great images. (It’s still not easy to make iconic images, mostly because billions of images are made daily, and it’s hard not to get lost in the crowd. But that issue is outside the scope this post.)

Fuji X-E2 / X-M1

The recently-released (November 2013) Fuji X-E2 is a mirrorless camera with interchangeable lenses (very well reviewed) that features an eye-level electronic viewfinder. Since images are not available, we display results for the X-M1, which has an identical 16 megapixel X-Trans sensor that features a new color filter array (CFA) pattern that reduces aliasing, allowing the anti-aliasing (Optical Lowpass) filter to be eliminated for improved resolution.

Fuji X-M1:  Normalized contrast plot

MTF is significant all the way out to the Nyquist frequency (right of the graph), which is consistent with the absence of an anti-aliasing filter. Sharpening is very conservative, with a maximum peak of 1.01. MTF values would improve with additional (small radius) sharpening. Noise reduction is also conservative, not starting until modulation drops under 0.1, which is lower than most of the cameras in this post. The New York Times just published an article about Fujifilm and the X-series, which has sold very well in a tough market.

Sigma DP1 Merrill

I won’t be purchasing this compact 14.8 megapixel camera because it has a fixed lens, and I want a zoom— available on several competitive cameras. I’m including it because I’m intrigued by the Foveon sensor, which has no color filter array and no anti-aliasing (Optical Lowpass) filter, and can potentially outperform Bayer sensors with comparable pixel count.

Sigma DP1 Merrill:  Normalized contrast plot

Moderate sharpening (maximum peak = 1.22) is not likely to be objectionable. MTF is outstanding for a 14.8 megapixel sensor, and it remains significant right out to the Nyquist frequency— a clear benefit to not having an anti-aliasing filter (and having a fine lens). The low amount of noise reduction is also notable: It doesn’t really start until modulation drops below 0.05 (1.16:1 contrast). This would contribute to Sigma’s reputation (attributed mostly to the Foveon sensor) of maintaining exquisite fine detail. Now, if only it had a zoom lens like the RX100…

Summary

To distill the above results, we have created a table of summary metrics.

• MTF50P, LW/PH, row 4.   The best overall sharpness metric for contrasty areas (where modulation ≈ 0.7). MTFnnP is less sensitive to oversharpening than MTFnn.
• MTF20P, LW/PH, row 4.   A metric that approximates vanishing resolution in contrasty areas.
• MTF50P, LW/PH, row 16.  The best sharpness metric for low-contrast areas (modulation ≈ 0.12).
• MTF20P, LW/PH, row 16.  A metric that approximates vanishing resolution in low-contrast areas.
• Peak (maximum)  A measure of the maximum sharpening MTF overshoot. Sharpening is excessive if too high.

Final comments and observations

All of the cameras discussed in this page make excellent images.

•  The 36-40 megapixel cameras (Sony, Pentax, Nikon) are capable of outstanding performance, but you won’t see it unless you
  • Use the finest lenses: Ordinary consumer-grade lenses can rarely take advantage of these very high resolution sensors.
  • Print large: At 13×19 inches (common for home printers) you won’t see a difference.
  • Use good technique: keep the camera steady during exposure and set the lens close to optimum aperture.

If you can’t meet all of these criteria, you can save a lot of money (and weight and storage) by sticking with a camera with 20 or fewer megapixels. 10-20 megapixel cameras with good optics are capable of making phenomenally fine 13×19 inch (or larger) prints, and they’re smaller, lighter, and less expensive. So why bother unless you want to look like a serious photographer (and maybe strain your neck in the process)?

• I’ve always been suspicious that mirrors compromise sharpness, especially for long lenses, though I haven’t done a controlled study. The lack of a mirror was one of the key advantages of the old Leicas (I still have my M2R), and it’s still an advantage, even though mirrors have improved (my EOS-6D is much smoother than my film SLRs). This makes the Sony A7R very attractive compared to the Nikon D800(E) and Pentax D645 (which is considerably more expensive and heavier). But I won’t run out to buy an A7R because few lenses are available (as of November 2013). They’ll be coming out over the next year, and I expect them to be excellent: I’ll be watching the reviews.
• Smaller mirrorless cameras (APS-C and Micro Four-Thirds from Panasonic, Sony, Olympus, Fuji, and others) are very appealing alternatives if you don’t need to make giant prints.
• Although the Panasonic LX7 is very good, the more expensive Sony RX100 II hits a real sweet spot when you balance its size, weight, and image quality. It’s a bargain for what it does. My son convinced two of his friends to buy them.
• Most of these cameras have fairly strong noise reduction that results in loss of fine texture. (The Pentax, Fuji, and Sigma are the exceptions.)  Whether or not this is a good thing is a matter of taste. You can recover the texture by turning down the noise reduction (if an adjustment is available) or by performing raw conversion off the camera using little or no noise reduction.

Sony has planted a bit of lust in my brain. I’ll be watching the growth of the A7R system. One day I may make a very rash, very large purchase (with a lot of thought and analysis behind it).

The post Sharpness and Texture Analysis using Log F‑Contrast from Imaging-Resource appeared first on Imatest.

This post describes how to measure color and grayscale patches on a variety of test charts, including Imatest SFRplus and eSFR ISO charts, the X-Rite Colorchecker, ISO-15729, ISO-14524, ChromaDuMonde CDM-28R, and many more, using a spectrophotometer and one of two software packages.

• Babelcolor PatchTool  which works with reflective test charts
• Robin Myers SpectraShop 4  which works with both reflective and transmissive (backlit) test charts

Measurement results are stored in CGATS files, which can be used as reference files for grayscale and color chart analysis in Multicharts, Multitest, Colorcheck, Stepchart, and SFRplus. In many cases, custom reference files provide more accurate results than the default values.

In addition, Argyll CMS is a free, Open Source, command line-based package that can be used for a number of measurements, including illumination intensity and spectrum. See the Argyll CMS documentation for more details.

Spectrophotometer

At Imatest we currently use the i1Basic Pro 2 (a good choice if you’re shopping for a new instrument; ). We’ll call it the i1Pro in the text below. It’s one of several instruments supported by Babelcolor PatchTool and Robin Myers SpectraShop 4 software. The i1Pro is primarily designed for reflective readings, but it can also be used to measure transmissive (backlit) charts.

• If you don’t have the original disk, Load the drivers and software from X-Rite. For our i1Pro we clicked on Software downloads (5 Items) (on this page), and loaded all relevant software. For the i1Basic Pro 2, go to this page.
• Install the drivers for your system and i1Diagnostics. Run i1Diagnostics to make sure your hardware is functioning properly.
• The other programs for the i1Pro are useful for purposes such as monitor calibration, but not directly relevant to chart calibration/patch measurement.

Babelcolor Patch Tool

Read the description of PatchTool on the Babelcolor page, then download and install it. You’ll need to purchase it in order to run Patch-Reader (no luck with evaluation mode). Registration is straightforward (no special instructions needed).

• Open PatchTool. The window is very simple— not much to it.
• Click Tools, Patch-Reader… to open the Patch-Reader…
• Select the Chat type. Click Custom chart for rectangular grid charts (like the X-Rite Colorchecker), and select the correct number of rows and columns. The default is 4 rows, 6 columns (for the Colorchecker). Select 4 rows, 5 columns for SFRplus color and grayscale charts. Unconstrained measurements work perfectly well for rectangular grid charts and must be used for charts that don’t have a regular rectangular grid (like the ISO-15739, CDM-28R, etc.). The Patch-Reader window will look like this.

Initial Patch-Reader window (4 rows, 5 columns)

•  Press Start. A window opens with the following message for calibrating the spectrophotometer. Follow instructions and press  OK  .

• After you finish the calibration, Read and Stop buttons appear on the lower-right of the Patch-Reader window.
• Place the i1Pro on the first patch of the chart and click the button on the side. This is equivalent to pressing Read (which should never be necessary).
• Now go through the patches one-by-one, by rows (for rectangular charts). For non-rectangular charts you’ll need to know the patch order, which is described below. For rectangular grid charts the Patch-Reader window will displays the location of the next patch to measure. Be sure to position the i1Pro carefully over the center of the patch and press it firmly against the target to minimize stray light. (Subdued room light doesn’t hurt.) (For a m row x n column chart, the order is Row 1-Col 1, Row 1-Col 2, … Row 1-Col m, Row 2-Col 1, etc.)

• You can use the mouse to select a patch to measure, i.e., you can go back and re-measure if you have any doubts about measurement quality.
• When you have finished the Patch-Reader window will look like this. You should examine this image carefully to be sure it resembles the chart— all patches should look correct:  You can still select patches to re-measure if necessary.

Patch-Reader window after all patches have been read

• Click Stop, and the a window appears with the following options.

• When you click Save, the following window appears. For typical Imatest use, only the L*a*b* data field needs to be checked. ID (SAMPLE_ID; the numeric sequence) and Name (SAMPLE_NAME; The alpha column and numeric row, i.e., A1, …, A4, B1, …, F4) may be checked, but will be ignored by Imatest. XYZ and xyY may be checked, but will be ignored if L*a*b* is checked. (The order of precedence in reading data is L*a*b*, XYZ, then xyY.)

• The save file name should include the media type (matte, luster, …) and any other information needed to identify the chart.

PatchTool writes files in CGATS format (an industry standard best explained in the PatchTool Help PDF document). Imatest can read CGATS files starting with August 2013 3.10 builds. Here is an example:

CGATS.17
ORIGINATOR    "BabelColor PatchTool, version 4.1.0 b321"
LGOROWLENGTH    4
CREATED    "2013-07-01"  # Time: 1:25:46 PM
INSTRUMENTATION    "i1Pro"
INSTRUMENT_SN    "608994"
MEASUREMENT_SOURCE    "Illumination=D50 ObserverAngle=2 WhiteBase=Abs Filter=UVcut"
ILLUMINATION_NAME    "D50"
OBSERVER_ANGLE    "2"
FILTER    "UVcut"
MEASUREMENT_CONDITION    "M2"
WEIGHTING_FUNCTION    "ILLUMINANT, D50"
WEIGHTING_FUNCTION    "OBSERVER, 2 degree"
KEYWORD    "DEVCALSTD"
DEVCALSTD    "XRGA"
#
# THIS FILE CONTAINS REFLECTANCE MEASUREMENTS MADE WITH PatchTool Patch-Reader
#
# Chart_type: "Custom chart"
# No_rows: 4
# No_columns: 5
# No_samples: 20
#
NUMBER_OF_FIELDS    3
BEGIN_DATA_FORMAT
LAB_L    LAB_A    LAB_B
END_DATA_FORMAT
NUMBER_OF_SETS    20
BEGIN_DATA
44.050    -4.479    -15.534
46.989    3.638    -16.591
38.129    6.395    6.321
52.539    -0.730    -3.700
40.766    32.354    14.498
42.772    -19.174    -21.994
57.309    12.668    12.012
53.507    1.771    3.460
47.533    -25.359    16.443
44.629    32.242    -6.113
61.667    -18.258    37.479
37.817    2.403    -27.741
33.075    4.416    -29.699
75.079    4.192    62.638
63.637    15.573    46.478
45.767    28.423    10.312
41.007    -8.057    10.701
59.762    -28.214    -1.841
55.213    23.810    34.126
33.661    10.341    -10.492
END_DATA

SpectraShop 4

Download and purchase SpectraShop 4 using links at the bottom the product description page. Both Windows and Mac versions are available. The brief Tutorials are recommended. You should also download the full PDF manual and look at the FAQ, which has troubleshooting information. Open SpectraShop 4. Click on the Measure specimens box, which is the second from the left. (I’ve tried Create and edit charts and Measure chart without luck.)

Measure specimens is the second box from the left This opens the Instrument Connection window, shown on the right. Select the spectrophotometer, click Connect, select the Specimen Type (reflective, transmissive, etc.). Be sure the spectrophotometer is on its base (with white tile), then press Calibrate. Instrument Connection window, shown just before Calibration
After Calibration is complete, the Measure Reflective Specimens page will open. None of the settings are critical, but we recommend Set Value: to 1. This is the number of the first patch. This corresponds to patch order better than the default 0. Set Root: to a short description of the chart or patch. Set Digits: to 2 (especially if you plan to measure over 10 patches). This will make the file sortable since the sequence will be 01, 02, …, 10, 11, … When you have finished making settings, press Start. Place the spectrophotometer sensor over the patch. If possible the room should be dimly lit to minimize light leakage. Take each reading by clicking the spectrophotometer button— waiting about two seconds for a beep indicating that the reading is complete. Patch order is described below. Take readings carefully. I haven’t (yet) found a convenient way to edit or redo erroneous readings. Saving the reference file I recommend saving results in two formats: the default SpectraShop 4 ss3 format (not interchangeable, but can be reopened in Spectrashop 4), which contains all results (including the spectrum), and CGATS with just the results used by Imatest. To save the SS3 file, click (in the window shown below) File, Save… or Save as… The procedure for saving Imatest-readable CGATS files is shown below.

Main SpectraShop 4 window showing Colorchecker results

To save the measurements in an Imatest-readable CGATS file, click File, Export… The window on the right opens. Select CGATS 1.7 ASCII as the Export Format. Spectrum should be unchecked and L*a*b* should be checked. XYZ and xyY may be checked if desired, but they will be ignored by Imatest if L*a*b* has been checked.   Spectrashop 4 Export window (for writing CGATS files)   Here are the results:

CGATS.17
ORIGINATOR    ""
CREATED    "2013-08-28 16:43:55"
INSTRUMENTATION    "i1Pro"
MEASUREMENT_GEOMETRY    "45/0"
MEASUREMENT_SOURCE    "A"
WEIGHTING_FUNCTION    "OBSERVER,2 degree"
WEIGHTING_FUNCTION    "ILLUMINANT,A"
NUMBER_OF_FIELDS    4
BEGIN_DATA_FORMAT
SAMPLE_ID    LAB_L    LAB_A    LAB_B     
END_DATA_FORMAT
NUMBER_OF_SETS    24 
BEGIN_DATA 
"Colorchecker-01"    39.80    14.43    17.09     
"Untitled-02"    67.86    22.51    20.21     
"Untitled-03"    48.51    -9.36    -23.27     
"Untitled-04"    43.04    -10.81    20.67     
"Untitled-05"    54.88    5.55    -24.67     
"Untitled-06"    68.04    -32.15    -6.00     
"Untitled-07"    66.70    34.88    63.83     
"Untitled-08"    38.48    -0.44    -45.06     
"Untitled-09"    56.15    47.12    25.25     
"Untitled-10"    31.91    17.66    -16.69     
"Untitled-11"    71.40    -16.98    50.80     
"Untitled-12"    74.84    21.00    71.57     
"Untitled-13"    27.04    0.66    -49.03     
"Untitled-14"    52.88    -32.25    23.97     
"Untitled-15"    47.07    55.90    37.55     
"Untitled-16"    83.99    9.76    78.29     
"Untitled-17"    55.75    47.38    -5.30     
"Untitled-18"    47.53    -31.32    -35.23     
"Untitled-19"    95.01    0.32    4.03     
"Untitled-20"    80.78    -0.72    0.38     
"Untitled-21"    66.36    -0.71    0.28     
"Untitled-22"    49.81    -1.43    -0.39     
"Untitled-23"    36.00    -0.85    -0.45     
"Untitled-24"    20.92    -0.34    -0.53     
END_DATA

Patch order

Charts with a regular m x n rectangular grid have patches numbered in a sequence shown in the example below for the 4 row X 6 column X-Rite Colorchecker. Use this order in acquiring patch data.

Several grayscale charts (ISO-15739, etc.) that don’t have regular m x n grids have their patch order listed in Special and ISO Charts. These charts are supported by Multicharts.

Patch order for charts supported by Multicharts can be displayed by running Multicharts, selecting 3. Split Colors: Reference/Input (in Display), then checking the Numbers checkbox. Here are the patch numbers for the DSC Labs CDM-28R.

Multicharts result, showing patch numbers, for the ChromaDuMonde 28R.

There are a few special cases, such as the Rezchecker (a tiny precision reflective chart for measuring color, tones, and MTF). The numbering follows the standard rectangular grid, but corner patches and patches used for slanted-edge MTF have L* = a* = b* = 0 in their reference file, and are omitted from color analysis. For the Rezchecker these include patches 1, 6, 8-11, 14-17, 37, and 42.	Rezchecker	Rezchecker numbering
Some cases are not obvious, like the eSFR ISO color patches, which are supported by Multicharts, but usually run in eSFR ISO. eSFR ISO color patches

See Also

• Color reference files in CSV or CGATS format
• Edmund Ronald’s Blog
• i1Basic Pro UVcut – predecessor to the i1Basic Pro2

The post Measuring Test Chart Patches with a Spectrophotometer appeared first on Imatest.

The post Microsoft Lync Logo selects Imatest appeared first on Imatest.

Details here .

Purchase here .

The post Infrared SFRplus test charts available now appeared first on Imatest.