Part 1: NikonScan

FACEBOOK_960x640pixels_sharpen_Congo_middle white_COPYRIGHT
Background information on the photograph:

After doing my military service as a paratrooper in what was then the Belgian Congo (fortunately there was no necessity for military action), I decided stay in the Congo and to travel until after the Independence, on June 30, 1960. Starting from Katanga Province I traveled north , visited the Rwenzori mountains and the volcanoes Mount Nyiragongo and Mount Nyamuragira in the Virunga mountains. I did a brief visit to Sudan, then I headed to Stanleyville, now Kisangani. From Stanleyville/Kisangani I decided to go to Ikela in Central Congo to assist to the Independence festivities, which presented drum playing and dancing on the local airstrip. I used Ektachrome film to photograph the colorful celebrations. A few days later a political crisis was started by the secessionist movements of Katanga and South Kasai Provinces. The pays became unsafe for Belgians. I rushed back to Kisangani, where I was taken prisoner by the Congolese army. Fortunaltely the Province Governor decided that I should go in the civilian jail. I was there for 10 days, not knowing what the next day would be. Fortunately on the 10th day the International Red Cross representative specializing in jailed, Mr. Senn, was able to free me. He accompanied me all the way to Kinshasa and stayed next to me until I was safely on a plane to Belgium.

The films, scanner and software:

But let´s  go back to pictures and film. I had started my trip with a supply of the original Kodachrome film (ASA 12). I guess I must have run out of Kodachrome because toward the end of the trip I used Ektachrome (ASA 32) (and perhaps Agfa or Gevaert?). After several years the Ektachromes had badly faded, acquiring a strong reddish hue.  For some reason that I don´t remember (maybe the Ektachromes had already started fading?) I had duplicates of a few pictures made on Kodachrome film by Kodak (according to Wikipedia Kodak had a low-contrast Kodachrome Commercial film for duplication). I only remember taking the train from Brussels to Antwerp where the Kodak lab was. Later on, after I moved to Brazil in 1978, I kept my transparencies in refrigerators, which stopped the fading. The original Kodachromes are still perfect.
I used the duplicates as reference for color restoration of the original Ektachromes, but even these Kodachrome duplicates have lost some color which can be seen in the greens which became brownish.
The transparency shown here was scanned with a Nikon Super Coolscan 5000 scanner; in this first part I will show and discuss the results using Nikon Scan 4.0 software; in Part II I will present the results with VueScan, which allows to work from the RAW file. In both cases I used infra-red cleaning.

Restoration options and trials:

I did first a straight scan to show the color shift; note that there is detail in the face of the man on the right.  I also did a straight scan of the Kodachrome duplicate to have a Faded - Kodachrome copyreference; but the  brownish greens indicate that it already suffered from some fading (possibly before I had it duplicated). The shadow details on the face are lost.

The NikonScan software includes Digital ROC, a plug-in for correction of fading which I tried at 2 values, 5 and 10 (there is little difference between the various levels of Digital Roc).

I first did the adjustments of the white point and black point for each of the color channels (R, B 72ppi_300pixels_sharpen_Nikonscan_without-dig-ROC_wh-pt-bl-pt-R-G-B_minus redand G) on the faded image. Often adjusting the white and black points of each channel centers the 3 histograms and gives a close approximation of the real colors. It was almost good, but the greeens remained somewhat brownish. Trying HUE didn´t help; I had to do a narrow lasso around the foliage, and within it SELECT COLOR RANGE to be able to remove some red from the leaves (not shown here).

Then I tried Digital ROC. I am showing here only DR 10.  You can notice a blue cast, but important 72PPI_D-Roc_Optimized_CROPis that now the foliage appears green, a blueish green but foliage is the only reference that is not  too subjective. This was easiest, and I think it gave me the best results:  I tried simply removing some blue and adjusting  green from the Digital ROC 10 image,  This last trial gave me true greens on the foliage. Sometimes you have to subjectively remember how the colors looked like (which is VERY subjective after more than 50 years!), green foliage can be a good reference (this is also valid for scanning negative film).

The Dibital ROC image is at left; at right is the corrected image. I arrived at practically the same result in two ways: LEVELS, Blue 63, Green 75, RGB 120, SATURATION 38; and LEVELS, Blue 87, Red 116, SATURATION 38 (shown).

Apparently the Digital ROC algorithm went deeper into the shadows in the RAW file than the straight scan, as more details can be seen on the shadow side of the face of the man to the right. Four-heads_TEXT_CURVES_CONTRAST-2

So this is what I got with NikonScan software. It may not have all the nuances of a new transparency (or a digital image) but at least the record of a unique event, in a region that is perhaps not accessible now, made it worth trying. In part II (for which you may have to wait a little!) I will attempt to show what can be done with VueScan software, which give us the RAW file to work with.


  1. Chris

    Hmm too bad there is no Part 2, will there ever be one? It was quite interesting to read the Part 1, as I purchased to Nikon 5000 ED from Japan with SF-210 and now have to start slide scanning 10.000 slides! 😀


    1. jjangoux Post author

      Hi Chris, I apologize for nor replying sooner. I was without internet for a month and a half after switching telephone/internet operators. Well, I don´t know when I will have time to do Part 2.
      Briefly, the problem is about scanning Kodachromes. NikonScan software gives double edges on high contrast edges if infrared cleaning is used (it also mistakes legitimate small black spots or thin black lines as dust or scratches); some claim that VueScan does a better job but it really doesn´t, it gives rough edges with infrared cleaning (many of my pictures have been with agencies and are VERY dirty). So what I did is use NikonScan, making two scans, an IR-cleaned one and a dirty one. I open both images in Photoshop, select on the dirty one the parts with double edge and copy them on a duplicate layer on the clean image. Of course I need to do some additional hand retouching. I decided that it´s probably better not to use IR cleaning and do hand retouching on the whole image but it´s very time consuming. You can also use Polaroid Dust and Scratch Removal tool on the scanned image but it works well only on uniform areas; it messes up if there are a lot of fine details. (I was told Polaroid doesn´t exist anymore as a manufacturing company but it sells its name). I decided not to use VueScan anymore as you can focus only on one point; with NikonScan you can focus on several points and make an average.
      Good luck with your 10.000 slides!



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s