Aerial Photo Processing Program/Beta Testers Sought

[tinker]

As a result of being somewhat unhappy with the status quo for photo handling, I have written a program that basically handles the process end-to-end.  It was suggested that some of you guys might be interested in helping with the testing.  All programs must have names; this one is called Phyllis.

Phyllis is designed to be run from a flash drive or memory (SD, CF, etc.) without any installation on the host machine.  The host machine must have Access and Excel available.  She automates four basic steps:

1. Photos are sucked off of the camera, rotated as necessary, and stored on the flash drive in a standardized folder structure based on the mission and sortie numbers.  These are the raw photos.

2. The raw photos are resized to the user's preference, renamed with a standardized name again based on the mission and sortie numbers, and stored in a second folder on the flash drive.  The sized photos are also captioned with a data block in the lower left corner that has the file name, date, and mission/sortie information.  This interim captioning is located so it will be overwritten by the final photo caption.  The original raw file dates are preserved.

3. The user is given an edit screen that allows him/her to directly edit the Objective, Lat/Long, etc. information for the photo and create an Excel caption file.  If the photo includes camera (EXIF) lat/long and/or  GPS time data, this can be automatically transferred to the caption file.  Phyllis' edit screen is fairly simple, so the user is also given the option to edit the caption file directly in Excel.

When the captions are complete, the user can print a report that basically mimics the WMIRS data entry screen, also showing the actual photo.  From this the user can do WMIRS entry and have WMIRS apply the captions.

4. Phyllis will use the Excel captions file to apply the captions locally in WMIRS format.  CAP logo, north arrow, etc.  These are again given standardized names and stored in a third folder on the flash drive.  Raw file dates are again preserved.

At all points, the photos can be reviewed full screen and also at full resolution for QC checking.  At all points, too, the photos can be downselected so that a subset of them are carried forward.

Ideally, a beta tester would have significant real-world experience processing mission photos and some familiarity with MS Access VBA so minor bugs would not bring him to a dead stop.  Please PM me with a direct email address if you are interested.  I am thinking that three or four people would be enough at this point.

Thanks in advance.

[wuzafuzz]

Sounds like a great idea to streamline processing.

Is Phyllis dependent upon any particular version of Access & Excel?

[tinker]

Is Phyllis dependent upon any particular version of Access & Excel?

Phyllis  was coded and tested with Office 2000 and (deliberately) with lowest-common-denominator controls.  Others have successfully run Phyllis on later versions.  One of the goals of beta testing is to find problems when trying to run on a variety of configurations.  AFIK no one has yet tried Phyllis on Vista.  So, we'll see.

[JoeTomasone]

I could use a program that:

1. Extracts, resizes, and renames the files from the camera.

2. Zips them along with a .gpx track file from another program (or grabs the file itself?)

3. FTPs it to a server.

We could make immediate use of that in FLWG for our RECON program.

[hatheier]

After going through a hours of pain-full post-exercise photo processing I really welcome this initiative and I would love to participate as beta tester.

I do have a number of photos, photo logs and GPS logs in my archive from prior SAREX's that we could run through the tool to ensure it helps in real-world scenarios.

[tinker]

After going through a hours of pain-full post-exercise photo processing ...

Yes, eliminating that is the goal.   Some of the North Dakota guys who processed 4,000 flood pix have been helping with the testing and they are quite excited.  PM me with your email address and a little background.

Extracts, resizes, and renames the files from the camera.

Easy.  Already happening.  Rotation is also available so the sized files are all right-side up.  An Excel file listing the camera file names and dates together with the standardized file names is automatically produced and saved with the sized files.  Normally, original file dates get blasted in this type of process but Phyllis restores them so the renamed files have the same time stamps as the raw files.

The photo paths look like:  J:\photos\09-T-9367\Air-03_Sized\09-T-9367(Air-03)S-07.jpg  One of the goals of the testing is to decide whether this is the right directory structure and file naming convention.

Zips them along with a .gpx track file from another program

Phyllis doesn't do that, but we are working with these guys:  http://photofinder.atpinc.com/how.html on some mods to their system that will result in lat/long, GPS time, altitude, and direction of travel being automatically added to the JPG EXIF (http://en.wikipedia.org/wiki/Exif) data.  Essentially, we'd be putting the GPS track info into the photos themselves.

From that, knowing the window the camera was shooting from, and doing some trigonometry on a photo series we are hoping to produce a good target position estimate.  All of that information can then be automatically inserted into the caption file for review and editing by the user.  Maybe that is another route to the objective you're seeking.

[Tubacap]

I would love to test it for you... this is a current topic of discussion here as well.

[Pumbaa]

I do 50 CD+ missions a season.  I run GPS while shooting.  I have beta tested many software packages, written 4 software manuals.  Expert in Excel and Access as well.

If there is a bug in the program I will break it in 30 seconds.

You need a beta tester on steroids.. gimee a shout.

[tinker]

Please PM me with a direct email address if you are interested.

[JoeTomasone]

Extracts, resizes, and renames the files from the camera.

Easy.  Already happening.  Rotation is also available so the sized files are all right-side up.  An Excel file listing the camera file names and dates together with the standardized file names is automatically produced and saved with the sized files.  Normally, original file dates get blasted in this type of process but Phyllis restores them so the renamed files have the same time stamps as the raw files.

The photo paths look like:  J:\photos\09-T-9367\Air-03_Sized\09-T-9367(Air-03)S-07.jpg  One of the goals of the testing is to decide whether this is the right directory structure and file naming convention.

That would work for me.

Zips them along with a .gpx track file from another program

Phyllis doesn't do that, but we are working with these guys:  http://photofinder.atpinc.com/how.html on some mods to their system that will result in lat/long, GPS time, altitude, and direction of travel being automatically added to the JPG EXIF (http://en.wikipedia.org/wiki/Exif) data.  Essentially, we'd be putting the GPS track info into the photos themselves.
[/quote]

That would be cool.   What I have now is a server backend that ingests a zip file with the images and GPX track log, resizing and geocoding the images, then overlaying them with the CAP MAJCOM logo and the pertinent details (lat/long/USNG/altitude, county name, and date/time).   Next step is to automate a Google Maps mashup. 

The images (at present) come out like this:



Resizing and geocoding on the PC side would save a lot of time for all concerned. 

[tinker]

geocoding the images

Are you writing EXIF data or just the visual captioning?  Do you do anything with altitude or direction of travel?

Resizing and geocoding on the PC side would save a lot of time for all concerned.

Well, we're pretty much there I think.  In addition to the captioning, Phyllis also maintains an Excel file with all the photo information so there is a kind of reference card available for human or machine.

I haven't mentioned it, but there is also an expert mode where the user can thoroughly confuse things by inventing his own directory structure.  More usefully, he can also choose what caption lines are included on the sized and final photos and whether the CAP logo and/or north arrow are added.  He can choose to show lat/long as DD MM.MM, DD.DDD, or DD MM SS to suit a customer's preference.

Next step is to automate a Google Maps mashup.

That's beyond my level of ambition right now but I'd be happy to work on making an easy interface.  Another thing that is beyond my level of ambition is a batch upload to WMIRS but it is certainly needed.

[JoeTomasone]

The images are both geocoded and have the data on them.  That's the way (uh-huh, uh-huh) they like it (uh-huh, uh-huh).

I will be determining the heading of the photo (fortunately, all of our photo windows are on the left side of the aircraft, so take the GPS heading and rotate -90 degrees...) and putting that in there as well.   There aren't any ambitions (that I know of) to do any further calculation with it.

The current mashup is at http://old.flwg.us/recon/ -- but it's still a very manual process.  The goal is to replicate (and slightly improve) this with a more automated procedure.

[tinker]

Thanks for all the responses. I will be contacting those who responded by the end of tomorrow. For now, the beta test program is fully subscribed and closed.