Photo/Image Database Management

C

ch1

Guest
I have now "digitized" my nearly 30-year collection of slides and print film. I also have a good work flow system in place for scanning new work.

All my digitized images are stored on both DVDs and to an external HD. Slides and print negatives are then being stored in a cool, dry, dark environment as "archives". However, the resulting problem now arises of how to manage this growing image database before it turns into a messy "rat's nest".

About a year ago I had a bried introduction to Picasa and do have it on some of my computers - but I haven't yet used it. While I think my own "organic memory" is the most "flexible" interms of relational management - it is probably the most "faulty" and prone to increasing failure! 😀

So, I am wondering what database management programs other folks here are using to organize their image collections?

I can divide my portfolio in several ways. I shoot both RF and SLR film (and, very rarely a film P&S) and, less often, both dSLR and P&S digital. I have gear in three locations (NYC, Copake and Tucson). With film I shoot both B&W and Color and with color I use both (chromes and print - the latter just for negatives).

I can probably come up with even more variables without thinking too hard!

However, given these parameters I'd be interested in both what kind of database software folks are using (and why) as well as their thoughts on how they organize their databases?
 
I've just started to use idImager, and it has a lot of features,. In particular, versioning, allowing different versions of the same photo to be "stacked", with only a single photo being displayed. Also, if for any reason you wish to use a different managment program, the database info can be exported.

The company is also very responsive to its users: I have an Epson R-D1, and (not suprisingly) was having trouble finding a image management program that recognised its raw files. After trying idImager, I emailed the company asking about R-D1 support - a day later they emailed me a beta with R-D1 support. Had to buy it after that!

Cons:

(1) The learning curve's a bit steep - I really must RTFM!
(2) It's a little slow, but each new version gets faster
 
JPG and TIF files have storage space available in a special area known as "EXIF" data.

http://www.exif.org/

Anyone who uses a digital camera already may be aware that the camera stores a great deal of information about the image in EXIF format on the image itself. Some photo and negative scanning software can be adjusted to insert EXIF information as well, although it does not know specifics such as aperture or shutter speed used, as a digital camera does.

This data that describes data is known as 'metadata' - data about data.

EXIF can also be used to store ownership information, copyright, permission-to-use information, keywords, and so on. There are many free-form fields as well as pre-defined fields that have specific meanings.

http://www.digicamhelp.com/what-is-exif/

A good image database management system will make extensive use of EXIF, which will allow the fields to be read, stored in a relational database, and indexed for quick searches. Additionally, an ideal image database management system will use XML as a lingua franca to export files and matching metadata to, for example, HTML format for the web - based on the results of search or query, etc.

Ultimately, one does not want to be locked into a system that requires all metadata to be re-entered, ignores existing EXIF data, and is proprietary - meaning that even if one can get their photos back out of the system, the metadata does not come with it.

EXIF is ideal in the sense that the metadata always stays with the image file, in addition to however else it might be used.

Good photo management software would allow manual categorization of such things as 'location' and 'date' wherever the information is not known from a digicam's files, and then allow searching on those parameters.

Examples I've used in the past and enjoyed include Thumbs Plus and ACDSee. However, I have not used them in several years, since leaving the world of MS-DOS / Windows behind and switching to Linux. I now do my own programming using PERL and the EXIF Perlmod.

http://www.cerious.com/photo-management-software.shtml

http://www.acdsee.com/products/acdsee/whatsnew

I believe that truly effective photo management software is still in its infancy, this is a growth industry, and only a few 'get it' with regard to transparency and adherence to standards such as EXIF. Most will read EXIF, a few will write EXIF, and very few will allow data extraction to db for indexing on EXIF. Overall, I still feel it is not soup yet.

Best Regards,

Bill Mattocks
 
Image Management for Dummies

Image Management for Dummies

I've tried a few image management systems and although they're very good, they take more care and feeding than I can keep up with. Based on the KISS principle, I devised a system of my own that requires minimal upkeep.

Image Management for Dummies:

My system begins with filenames. In fact they're the heart of the system. I find it very important that images be displayed in chronological order, so every filename begins with a date stamp (approx date if exact date unknown) and a sequence number. e.g.,

20041215-035g--sunrise-harbour-port-credit.jpg (or .tiff or .crw etc.)
20050600-b2-24--marion-coffeeshop-oakville.jpg

If I don't have a day, I insert two zeros in the day field so that the filenames sort properly. Example 1 is for digital images. After the date is a hyphen, then the digital sequence number of that day followed by my first initial so I know I took the shot. If my wife took the shot, I add m instead of g.

The second example is for scanned film. It tells me that this is frame 24 of the 2nd B&W roll I developed in June, 2005. I use b for B&W and c for col and start a new roll sequence each month. My stored negatives in Printfile sheets are marked with the same base name: e.g., 20050600-b2 and are also stored chronologically.

Next come the descriptors, after a double hyphen. Usually something about the subject and the location. File names can be up to 64 characters long, so there's room for descriptors. Makes for long filenames, of course, but that doesn't matter. I personally try to avoid putting spaces into filenames and I keep everything lower case.

In Windows or Linux I create directories (folders) by year -- 1964, 1965 ... 2005 etc and move the appropriate images to the appropriate folders.

To find things I use Windows Explorer Search or the Unix find command to look for keywords.

Here's a link to a fuller explanation: http://www.northernjourney.com/photo/articles/filenaming.html

It takes a little up-front work, but then I'm done. It is non-proprietary and can be moved from computer to computer and OS to OS easily. So far I've always been able to find any photo I was looking for, and I have many thousands of digitized images.

It makes backup fairly straightforward too. The reason for the double hyphen before the description is that I have some Perl scripts that allow me to make an index.html page on my CD-R backups, and that allow me to strip off the descriptors when I need shorter filenames on copies of the images.

Not fancy, quite simple once you get the discipline, and for me at least, it works!

Gene
 
I ran through the three-week trial of iView Media Pro. I like it.

But I'm going to try the cheaper 'not Pro' version before I pay any money for it...

I need something. I tried just using folders and filenames, but it wasn't working for me.
 
Bill is corrrect, of course, about the non-existence of an ideal photo database system (or, at least, one most can agree on). I'm tentatively planning to use a Filemaker-based system for my database; not perfect, but it will allow me to keep better track of stuff as I scan and collate. Also looking into apps by Extensis/CreativePro, iView and others, but since I already own Filemaker, it obviously gets the first nod.


- Barrett
 
amateriat said:
Bill is corrrect, of course, about the non-existence of an ideal photo database system (or, at least, one most can agree on). I'm tentatively planning to use a Filemaker-based system for my database; not perfect, but it will allow me to keep better track of stuff as I scan and collate. Also looking into apps by Extensis/CreativePro, iView and others, but since I already own Filemaker, it obviously gets the first nod.

- Barrett

What is interesting to me is that all the tools are there - but no one seems to be sticking them together correctly yet!

EXIF, XML, and a nice relational db with search and export capability in a variety of formats, such as burning to CD and/or ftp'ing to a website as HTML. Perhaps a wizard to help new users categorize their shots the first time through. This is not rocket science and should not be this hard.

Best Regards,

Bill Mattocks
 
bmattock said:
This is not rocket science and should not be this hard.
Write one?

There's got to be enough in the way of capable coders hanging around RFF.

I'm tempted myself, but:

1) I don't need another project.

2) I have much more experience in web apps - I could build a server-based thing, but that's be of limited use to anyone.

3) I don't have time for another project.

4) But all that said, I'm sure I'd be tempted to contribute.

Forget image processing functions - they're not necessary - but a sensible database structure and interface, all the other shiny bits like outputting a static HTML gallery... it's just time and effort. As you say, Bill, it's not rocket science.
 
Last edited:
I created a simple database in Filemaker Pro (mac) based on the included photo database. I modified so that it would import location and references to all my photo cds and I also made fields to track where I am posting the images. It does not handle exif or iptc, however. There is a pluging for such support but it is pricey and requires some scripting. On of these days I'll get around to it.

There are some php gallery programs that handle exif/php and it would be interesting to see a php program dedicated to photo mgt.
 
Back
Top Bottom