fogbound.net




Mon, 16 Sep 2013

Limited Comparison of Negative/Slide Scanning Services

— SjG @ 9:42 am

I take too many pictures (a flaw that I have had for several decades). I am currently trying to aggregate my photo library into one single digital repository. This required sorting through tens of thousands of negatives and thousands of slides, and identifying the ones I want to digitize. Then, after the first pass of this operation was complete, the next step is to find a commercial scanning service1.

After doing some research, and setting some criteria, I came up with several possible services. I built a feature/price matrix based on information they published on their web sites (current as of July ’13). I then decided to send some samples to each service.

To best represent the kind of stuff I’d be having scanned, I decided I would send:

  • a sheet of roughly 25 B&W negatives (all from the same trip, thus the same film stock, generally same exposure range, generally similar condition)
  • a sheet of roughly 25 color negatives (all from the same trip, some of which had been cut into nonstandard strip lengths)
  • a collection of 20 slides, including Kodachrome, Ectachrome, and unknown slide stock

The process of doing this narrowed down the seven to five. I ruled out scancafe.com because they do their scanning overseas, and I’m a raving xenophobe. Well, that, and the turn around time seems to be significantly longer for them. I also removed fotobridge.com from the running because they didn’t have any way of doing small batches; using a service for “up to 250 slides,” say, when I only needed 20 didn’t make sense.

I got my first batch back from ScanMyPhotos a staggeringly fast 3 days after I’d shipped out the negatives. The majority of the others came in after about two weeks, with the last set from ScanDigital straggling in last after a little more than one month.

ScanMyPhotos.com
ScanMyPhotos sent back two DVDs. DVD “Untitled”. Three directories, labeled “00000001,” “00000002,” and “00000003.” Images named for directory with an index, e.g., “000000010001.jpg”. The Exif reflects scanning by “EZ Controller.”

Note that the pricing average for ScanMyPhotos in the conclusion spreadsheet is skewed if you’re not scanning slides. If you’re just scanning negatives, ScanMyPhotos is significantly less expensive than the others.

B&W scans are nicely contrasty, and nearly free of dust and specs. They have a visible “dither” noise, which looks like the results of an over-aggressive sharpening algorithm; running Aperture’s noise-reduction with gentle settings is a big improvement. The color negatives had good saturation and detail. The slides also look good to me, except for a strange grid that is visible in the darker images which is not apparent in the slide itself.

B&W

B&W

Color Neg

Color Neg

Slide

Slide

Dark Slide w/ Grid Effect

Dark Slide w/ Grid Effect

DigitalMemoriesOnline.net
Digital Memories provided a CD in a nice DVD short pack, labeled (externally) according to the project name I provided them, while the volume name was “New”. It contained three directories, named based on what had been written on the top of each negative sheet. Within each directory, the files were named for the directory with an index, e.g., CuscoTamboPiscalIMG001.jpg. Exif reveals they were scanned with a Nikon Super Coolscan 5000ED.

B&W images were bright and contrast was good, but there’s a fair amount of dust and scratches visible. Color negatives are also good, although it seems that both red and green saturation are slightly higher than they should be. Slides were quite good.

B&W

B&W

Color Negative

Color Negative

Slide

Slide

DigMyPics.com
DigMyPics sent back a CD labeled “493722_CDJPEG1” in a nice DVD short pack. It contained three directories, named based on what had been written on the top of each negative sheet. Within each directory, the files were named for the directory with an index, e.g., Machu_Picchu_35mm_001.JPG. Exif identifies the scanner as a Nikon Super Coolscan 9000ED.

Pricing for DigMyPics may be a bit misleading, as you can select online to reject up to 20% of the pictures after they’re scanned, and you don’t pay for them. You can view them on the web to choose which (if any) to reject. Only now when I write this, I realize that I don’t recall whether the selections are full resolution. I can only imagine that they’re not, otherwise you could download them without paying. In any case, don’t expect to select your rejects based on pixel-peeping.

B&W were nice and mostly contrasty, although the dust and scratches were still plentiful. Color negatives were good, also showing some of that slightly heavy red/green saturation. Slides were pretty good.

B&W

B&W

Color Negative

Color Negative

Slide

Slide

Southtree.com
Southtree sent back the negative sheets folded at the bottom of a box stuffed with paper. The DVD was labeled “Goldstein,” but the picture file names all inexplicably started with “Godstein.” Unlike the other services, the three kinds of images were all saved into the root directory of the DVD, with an index, e.g., “Godstein_3_037.jpg”. Exif did not contain any scanner information.

B&W were good and contrasty, but had a lot of dust and scratches. Color negatives had good color, but showed a surprising amount dust and scratches. Slides has good color, but they are all out of focus, even for negatives that are quite sharp.

B&W

B&W

Color Negative

Color Negative

Slide

Slide

ScanDigital.com
ScanDigital provided a DVD in a standard jewel-box, labeled (externally) with my name and order number. The CD volume name was “My Photos” and it contained two directories, “35mm Negatives” and “Slides”. Within “35mm Negatives” there were two further directories, “1” for B&W negatives and “Thailand” for the color negatives. Filenames reflected order number, folder, and index, e.g., 110798-35mmNEG-Thai-001.jpg for a color negative, and 110798-SL-001.jpg for a slide. Exif calls out the scanner as a Nikon LS-5000.

B&W have poor contrast, although the dust and scratches is better than most. Color negatives have muted colors, and not only show scratches but also have disturbing compression artifacts in gradient areas. Slides, however, are excellent both in color and sharpness.

B&W

B&W

Color Negative

Color Negative

Slide

Slide

Caveats
A comparison like this is fraught with opportunities to be unfair. I’ve tried as much as possible to separate the objective and subjective judgments so that you can draw your own conclusions. There’s also a whole lot of details, some technical, some not, that go largely unstated above. So in the interest of trying to represent everything as accurately as possible, here are some disclaimers:

I went for the low-end “consumer grade” processing with all of these services. It is quite possible that the more expensive offerings, targeted more at pros, would have better results.

Dust and scratches are always a challenge, and may not occur the same on all exposures (even on the same roll of film). Digital ICE, the dust and noise reducing technology used by Nikon scanners, does not work with B&W film and some kinds of slide film.

I sent in my images with no special notes, requests, or instructions. Some services have a space on their forms, which may (or may not) allow you to be specific about special handling (e.g., do not sharpen, or please make high contrast, etc.)

Conclusions
Attached here is a downloadable spreadsheet with both the objective and subjective rankings of these services.

1 True, I could scan them all myself. However, at 2-3 minutes per scan, it’d take me hundreds of hours that I could otherwise spend cluttering my collection by taking more pictures.


Mon, 2 Sep 2013

Legacy-labeler.com

— SjG @ 5:29 pm

Consider this the soft-launch announcement for Legacy Labeler, an experimental web site for adding metadata and labels to your family pictures.

Samuel

In Ye Olden Days, family pictures would be physical photo prints all stuffed in a drawer somewhere. It’s possible that they were neatly labeled and put in albums, or mayhaps someone wrote names on the backs.

Today, however, we are deluged in JPEGs straight out of our digital cameras. Some people have discipline and create digital galleries, but more often they’re just accumulated on hard-drives somewhere.

Legacy-labeler.com is an approach for labeling these JPEG images, and adding or updating the relevant metadata in the file. Future generations deserve to know more about the pictures than when it was taken. “DSCN1012.jpg” probably won’t mean a whole lot to them.

I welcome feedback on the site: suggestions, bug reports, comments, flattery, invective, snark, or what-have-you.


Sat, 29 Jun 2013

8-bit Paleontology

— SjG @ 9:28 am

My maternal grandfather was a paleontologist of some note, yet it was no small surprise that L— broached the subject upon the occasion of my recent visit.

L—, a colleague from former heady days in the dot-com era and now a Web 2.0 millionaire in her own right, has used some of her new-found wealth to indulge in her original academic interest, the study of 8-bit Paleontology. In what was a rare honor, L— permitted me to view her private collection.

This collection, housed in the Library of her estate, is a well-lit room, dominated by the mounted skeleton of a contemporary Homocongregus namco. Surrounding this modern skeleton, she has a splendid display of fossils of some of the great precursor species, including Homocongregus dotovorus and Homocongregus arcada.

Today, however, she was pleased to show her latest acquisition. Opening a velvet-lined box, she revealed to me what is arguably the world’s best example of the immediate ancestor to Homocongregus namco: a gorgeous fossil of Homocongregus Phasmaprosecutus preserved in stone for all eternity in the very act of consuming a row of equally well preserved Pacdotus powerupi. It is truly a stunning piece, worthy of any great museum, and incontrovertible physical record of the evolutionary path of the species.

She kindly allowed me to photograph this specimen, thus I am able to present this image to share with the world, and contribute to the greater good of science:
Homocongregus Phasmaprosecutus


Thu, 13 Jun 2013

Failures in Image Processing

— SjG @ 10:00 am

I have a tendency to have ideas that seem simple until I attempt the implementation. Then, all of the complexities jump out at me, and I give up and move on to the next idea. Once in a while, however, I’ll go so far as to prototype something or play around before abandoning ship. This here’s an example of the latter.

The Big Idea hit when I looking up a recruiter who had emailed me out of the blue. He shared the name of someone I went to school with, and I wanted to see if it was, in fact, the same person. In this case, a quick Google Images search on the name and job title indicated that it was not the same person, so I didn’t have to fake remembered camaraderie.

While searching, though, I thought it interesting the variety of faces that showed up for that name. Hm. Wouldn’t it be cool, thought I, if I could enter a name into my simple little web service, and it would spit back the “average face” of the first ten or twenty matching images from Google? After writing a few pages of code, scrapping them, switching languages and libraries, writing a few pages of code, scrapping them, switching languages and libraries again, writing a few more pages of code, I then ditched the whole enterprise.

In the process, I did use some morphing software to test the underlying concept. Here is the average face for my picture mixed in with the first seven Samuel Goldstein images that were clear enough, angled correctly, and of sufficient size to work with:
s0123_4567

For what it’s worth, here are a few of the software challenges to automating something like this:

  • Extracting the portrait from the background. This isn’t critical, but will simplify subsequent tasks.
  • Scaling the heads to be the same size.
  • Aligning the images more or less. I was going to use the eyes as the critical alignment points; if they couldn’t be within a certain degree of accuracy, this would suggest incompatible images (e.g., one a 3/4 portrait, the other straight on).
  • Detail extraction. This is finding key points that match on each image. Experimentally, when combining images by hand, it may be sufficient to match:
    • 4 points for each eye, marking the boundaries
    • 3 points marking the extremities of the nostrils and bottom center of septum
    • 5 points marking ear boundaries: top point where they connect to the head, top of extension, center, bottom of lobe, point where lobe connects to head
    • 5 points marking top outer edges, outer angle, and center of mandible
    • 5 points mapping hairline
    • 5 points mapping top of hair
    • 7 points along the superciliary ridge
  • Interpolate these points on each pair of images, then on each interpolated pair, and so on until a single final interpolation is achieved
  • Render the image to a useful format
  • View image, and laugh (or cry)

A few other lessons learned:

The typical picture returned by Google images search for a name will be a thumbnail portrait. It’ll be small — maybe 150 by 300 pixels or so. While that’s enough data to recognize a face, it’s not a lot of data when you start manipulating it. Ideally, for nicer results, source images should be a minimum of two or three times that size.

Google gives back different results depending on whether you surround the name with quotes or not; it also makes a big difference if you pass size or color parameters. The “face” search is a good parameter to use, although when searching for “Samuel Goldstein” face search inexplicably had lots of Hitlers and Osama Bin Ladens. The “safe search” feature is also strongly recommended for this project — again, when searching for “Samuel Goldstein” without safe search yielded a variety of unexpected vaginas.

Bing image search gives different results (not too surprisingly), but they also have some anomalies. My search brought back many of the same pictures, along with an inexplicable collection of calabashes and LP labels.

If any ambitious programmers go ahead and implement this, please send me the link when you’re done!


Mon, 8 Apr 2013

Why I need an extension on my taxes

— SjG @ 9:49 pm

… because I wanted to play with HTML 5 Canvas.
example1

example2
This should run on any reasonably sane browser. Play with it yourself here.