Measuring Relative Film Density Using A Scanner

Introduction

To get the most from traditional black and white film it is often necessary to determine how best to expose and develop specific film-developer combinations by making film tests. Methods for making such tests are well described on other web sites. A good example is http://www.zone2tone.co.uk/testingm.htm#film_speed . A reliable method of measuring density relative to base and fog is (pretty much) essential for testing but unfortunately most of us don't have a dedicated densitometer easily to hand. I have discovered that my flatbed scanner (Epson Perfection 3490) makes a reasonable substitute. I suspect most other modern transparency scanners will also give satisfactory results.

Although this method gives very repeatable results it does not give exactly the same readings as standard densitometers. A friendly member of Ilford’s on line forum named “Engbob” was kind enough to cross check my film test samples using his densitometer. My measurements of relative density for areas around “0.1 D above base and fog” came out 15% lower than measurements made with the “real” densitometer. I believe the discrepancy between readings is a consequence of the Callier effect. Standard densitometers illuminate film from a point source and consequently measure specula density. Most scanners have diffuse light sources and consequently measure simple diffuse density. Unfortunately there is little that can be done about this problem other than make some rule of thumb allowance.

Despite the shortcomings I find this a good method point for testing B&W film in the absence of a standard densitometer. This web page explains the process I use.

Fixing The Scanner Exposure Settings

Set the scanner to record a B&W negative. On my scanner this equates to “B&W negative film” and “8 bit greyscale”. It is necessary to ensure the exposure settings are fixed! I know of three ways this can be achieved but I only recommend methods one & two, the third method should only be used as a last resort.

(Method One) Simply use the "Preview scan" if these scans are made before automatic exposure adjustments. Fortunately this is the case with my Epson scanner.

(Method Two) Manually override the scanners automatic exposure adjustments. Some scan software has a user interface that allows you to enter your own exposure values. Make sure you always enter the same values every time and make sure you haven't missed any out! (This is a little fiddly in practice but gives sound results). The screen-dump below shows settings used with my scanner. Please note the configuration of the user interface will look different with different types of scan software!

Manual exposure settings

(Method Three) Force the scanners automatic exposure settings to have the widest possible range of densities in the scan area. One area in the scan should be completely transparent e.g. not covered by any film at all so the scanner sensor looks straight at the light source. Another area in the scan should be completely opaque e.g. thick black card. Both of these reference areas should be close as possible to the scanned film sample. An example is shown below. (This method is fiddly and may give slightly unreliable results)

Force Scan Settings

Recording The Scanner Output Values

I know of two methods to record the scanner output values. With both methods it is best to make a warm-up scan first to give everything inside the scanner a chance to get up to a working temperature, especially the lamp. Discard this first scan and then make a second scan shortly afterwards for taking measurements from.

(Method One) The software that came with my Epson scanner has a tool called densitometer which allows me to take readings from pixels shown on the preview scan. If you are using similar Epson Scan software move the crosshairs over the point on the negative that you wish to measure and then use the reading labelled "Before" shown circled in the screen-shot below. (This is by far the easiest method)

Epson Densitometer Tool

(Method Two) After fixing the scanners exposure settings, scan the film sample and import it into photo editing software such as Photoshop. Convert the scan into a grey scale image if this is not already the case. (Some people advocate doing a colour scan and then measuring the blue channel however). Sample the image with the eye dropper tool and then take readings from the foreground Colour Picker window. The R, G & B readings should all give the same value so take any one of these. The screen-dump below shows the view in Photoshop. (Although this is a fiddle I find it gives the identical results to my first method)

Photoshop Eye Dropper & Colour Picker

Calibrating The Scanner Output

Purchase a Stouffer 21 step wedge (I got mine from an American web site for 11$ http://www.capefearpress.com/order.html ). It's a piece of film material with 21 individual density steps. Every density increment is 0.15 density above its predecessor. Measure the scanners output value at each step using the previously described methods. You should end up with a scan & table of values like the ones show below.

Step Wedge Scan

Calibration Data

Step No.

01

02

03

04

05

06

07

08

09

10

11

12

13

14

15

16

17

18

19

20

21

Scan Val

16

59

95

124

149

169

184

197

207

214

222

227

233

236

239

242

243

245

243

246

246

It is necessary to calculate a conversion function to turn the scanner output values into sensible density readings. When scanning film as negatives I find the following function is best for this conversion.

D = a – log[base b] (c – L)

Where

D is the log film density

L is the value read from the scan

a, b & c are constant values chosen to suit the scanner being used.

To find values a, b & c I have made up a spread sheet to make the task as easy as possible

Negative_Densities.xls

Negative_Densities.ods

Adjust the coefficients to get results showing 0.15 density increments between the wedge steps. Unless you use a very fancy and expensive scanner, the density range of the step wedge will exceed the range that the scanner can accurately measure. It is only necessary to consider the steps 1 to 15 because steps above 15 have densities greater than 2 which is outside the range normally used in B&W film, we can safely ignore these step values. The measured densities should plot to give a straight line up to density values of 2. The plot below shows the results from my own calibration.



Plot Of Calibration Densities

Measuring Relative Density

Using the previously described methods, measure the scanners output values given for film base and fog and for the area of interest on the negative. Scan both areas of film in a single scan, do not scan them separately or the results will be less accurate! Calculate density values for these two points using the formula and coefficients derived from calibration. Subtract the first result from the second to give relative density. I find it convenient to make this calculation on the same spreadsheet used for calibration. It's far easier than trying to enter the formula into a calculator. Both spreadsheets TheoreticleDensityCalc.xls and PracticleDensityCalc.xls make the calculation as simple as possible. The screen-shot below shows the sheet being used for calculating relative density. Scanner outputs are entered into the light blue cells.

Spreadsheet Screen-Dump

Conclusion

Measuring relative film density using a transparency scanner is slightly more hassle than using a purpose made densitometer. However if you haven't got your own densitometer but are serious about doing occasional Zone style film tests then I think its well worth the effort. The cost of the step wedge is negligible compared with buying a densitometer and these make a useful reference anyway. There are several academic papers available on the internet (you have to pay to download them though) these focus on using transparency scanners to evaluate film densities in medical radiology. The scanners used are usually fairly normal if tending towards the high end of what is commercially available. Their use seems quite common place nowadays in many similar medical applications. This gives me a confident impression that commercially available scanners are reliable and accurate enough for scientific film measurement. I think the techniques I have outlined are basically sound and reliable and I hope they prove useful to other amateur B&W photographers.

Back to main photography page