Showing posts with label software. Show all posts
Showing posts with label software. Show all posts

Saturday, July 7, 2012

Another Doggy Picture


This is actually the first file out of the new camera (see last post). 

This is the first one I have presented that is finished in Lightroom 4.  I probably wouldn't have upgraded Lightroom for a while, except the the D800 raw files are not supported in version 3.  However, I was surprised to find version 4 is actually a pretty useful upgrade.  Most of the controls that work globally on an image can now work with most of the local adjustments too.  Adobe has given us access to the individual channels in the tone curve tool, and added a few new modules ( book maybe the most useful?).  It is worth the upgrade cost.  Heck, it is worth the $150 for the full version too (version 3 cost $300!!!).  I'm still debating on the Photoshop upgrade.  I think CS5 can read the D800 files, and even if it can't Lightroom render them prior to passing it off to Photoshop.  I rarely dump anything into Photoshop, so I may just leave it alone for now.  Anyone who uses CS6 care to comment?

Anyway, enjoy the doggy photo.  I promise this will be the last post for a while that deals with tools.

Wednesday, April 6, 2011

Scripting Photoshop


A long, long while back I talked about scripting the Gimp.  At the time I knew Photoshop allowed the recording of Actions.  What I didn't know is that the Gimp script I wrote can basically be done in "real time" using Lightroom's Detail panel-- better, and faster, and with visual feedback.

I also I didn't know Photoshop has a full-blown scripting engine in it too.  The scripts open into an editor titled ExtendScript Toolkit CS5.  The language used is Javascript.  For me its a heckuva lot easier to read and code than the Gimp's Scheme.  To be fair, I think Gimp can use Javascript too.  I just didn't find too many examples of it.

Why am I writing about this?   I need to quickly piece together a book, and I ended up coding to make it happen.  I had each page of the book loaded as a layer, and I wanted each layer pasted over the background layer and saved to its own file.  Photoshop ships with a script called "Export Layers to Files".  I was so excited!  But, alas, it didn't quite do what I wanted.  It would not include the Background layer.  I did a small amount of digging and found out how to edit the script to force the Background layer to be present also.

However, this wasn't good enough for me in the long run.  I'd like to be able to toggle the ability to save the background layer or revert to the original functionality of script.  I spent a few moments tonight to tweak the user interface, and add ability to the script.  Here is a screen grab of my tweaked dialogue with my new option selected.


If you want to try this for yourself, below are my tweaks to the code... I included some of the original lines of the file so you can find the place in the script to make the changes.  I don't think I can distribute Adobe's code, and I don't have diff installed on this computer, so hopefully this will help :

This is found under the global variable section (note, these are getting split into two lines, they are 1 line each, be careful of linebreaks through these snippets -- cutting and pasting them into the editor seems to do the "right" thing.)

var strCheckboxVisibleOnly = localize("$$$/JavaScripts/ExportLayersToFiles/VisibleOnly=&Visible Layers Only");
// KT alterations
var strCheckboxIncludeBackgroundLayer= localize("$$$/JavaScripts/ExportLayersToFiles/IncludeBackgroundLayer=&Include Background Layer");


This is in the function settingDialog

// -- the fifth line in the dialog
dlgMain.cbVisible = dlgMain.grpTopLeft.add("checkbox", undefined, strCheckboxVisibleOnly);
dlgMain.cbVisible.value = exportInfo.visibleOnly;

// -- adding my include background layer checkbox KT
dlgMain.cbIncludeBackground = dlgMain.grpTopLeft.add("checkbox", undefined, strCheckboxIncludeBackgroundLayer);
dlgMain.cbIncludeBackground.value = exportInfo.includeBackground;


This is in the same function, my code is the second line, remember the first is to help you locate it.

exportInfo.visibleOnly = dlgMain.cbVisible.value;
exportInfo.includeBackground = dlgMain.cbIncludeBackground.value;


In the function initExportInfo I set the variable to false

exportInfo.includeBackground = false;


And finally in the function "main" right before the call for exportChildren I inserted this block of code:


if (exportInfo.includeBackground == true){
try { duppedDocument.backgroundLayer.visible = true; }
catch (e) {} // Background layer is kept. In future add this as a dialogue?
}


You can see my comment is a remnant of my first pass at making this work. Programmers leave comments in code long after they are useful, so I felt no need to clean it up.

The original Adobe script is in the CS5 install directory:

\Path\To\Adobe\Adobe Photoshop CS5 (64 Bit)\Presets\Scripts

Wednesday, July 28, 2010

Vuescan


I recently finished up a roll of Kodachrome and had it developed.  This got me thinking (again) how I was going to get my slides and negatives from my photographic beginning into a digital form.  I want to do this partly for backup purposes, and partly because it is much more likely I will see them if they are in my Lightroom catalog.

This is of particular importance because all the pictures I have of Kathryn until she was about 4 years old are on film!  The photo at the top was from way back when Kathryn was a little beaner.

I have an older film scanner, but it wasn't supported by Windows 7.  I found a program called Vuescan.  Not only does it work, but the scans seem better than what I used to get out of original software.

However, I really don't like all the dust and scratches that come with film.  Not to mention the scanning itself is ridiculously time-consuming.  There is a reason no one with sense uses film anymore.  I am only somewhat kidding.  However, I have decided I will at least play with it again.  The Kodachrome shooting was a fun diversion.

I never claimed to have any sense.

Monday, June 28, 2010

Ghetto Processing No More


I did it.  I couldn't take it any more.

I broke down and bought some real software to work with my images.  I had been avoiding it for a very long time.  It was one of things I knew I would do when I absolutely had to, but its an expensive endeavor.

I reached the point I had to.

My previous workflow of Nikon Capture -> ViewNX (white balance, camera mode) [to jpg] -> Gimp ( sharpening, any fixing [ not much ], any cropping ) -> [final jpg] worked fine when my projects required only 10-12 finished images.


However, lately the amount of finished pictures from a project was not stopping at only 10.  I was finishing, in some cases, over 100 images!  My workflow broke down.  ViewNX is slow, and will not show you changes in real time.  Gimp, is also pretty slow, even though I had a lot of items scripted and batch-scripted to work on complete sets of images, it was taking way too much time.

Lightroom fixes that issue.  I still use Nikon Capture to import images off my cards, but then I take them right into Lightroom.  I can do almost everything in Lightroom, very quickly.  For anything I need to do that Lightroom cannot, I now have Photoshop.   I am really surprised at how much nicer Photoshop is to use than Gimp.  I didn't expect that to be the case -- but, wow, it is a lot easier to work with.

I think my original workflow worked pretty well for a long time, but I resisted switching too long.  I burned up hours of my life.

There are a bunch of other benefits to using "industry" standard tools, but for me, the most important one is extra time.

(funny aside... my original intention here was to place a flash and radio trigger under the water and bubbles, but I chickened out.)

Saturday, March 13, 2010

Christmas in March


We celebrate Christmas with a good friend of ours far removed from the actual holiday.  There is just too much going on.  Its so much more fun to do this way.

When we got together today it reminded me that I had some frames I made with the intention of making a garish HDR image of our Christmas tree.  When I tried to do that in December I found the software I was using, Qtpfsgui, was clipping highlights and filling them with black.  This was far from desirable.

I decided to look for some other software.  I found Luminance HDR, which is a successor to Qtpfsgui.  I was disappointed to find the bug I encountered earlier was still there.

I dug around a bit more on the net to see what I could find.  I came across Picturenaut.  I found this program to be quite fast at creating the HDR image.  It also was not clipping highlights!  However, I was disappointed in the tone mapping options and output.  Qtpfsgui really spoiled me with a great toolbox of over the top tone mapping options.

I wondered a bit about what to do.  I had a program that tone mapped the way I wanted, but couldn't create the initial HDR image properly.  I also had a program that created the HDR image correctly and quickly, but couldn't tone map the way I wanted*.  Fortunately they both supported OpenHDR file format!  I created my HDR file with Picturenaut and then imported it in Luminance HDR for final tone mapping.

The final image was made by creating a few different tone mapped jpeg files and importing them into Gimp as layers.  I simply changed the blend mode to "screen" and played with the opacity of each layer until I got a final look I liked.  It's just what a wanted, a slightly crazed "record" shot of our Christmas tree.

Now that I got my image out I will revisit Picturenaut a bit more.  There are plug-ins available for it that may give me the tone mapping options I want.  I also was very impatient with it today.  I knew what the algorithms did that are supplied in Luminance HDR.  I was less sure about where I wanted to go in Picturenaut.  But, it works SO quickly on my machine I'd like to make it my primary tool if possible.

If you are interested in HDR imaging take a look at the programs I listed above.  The commercial program Photomatix is the gold standard piece of software to perform this process.  If you wish to do a lot of tone mapped images it is probably the route to go.

*Or, more likely, I am not familiar enough with it yet.

Thursday, February 18, 2010

Timelapse video experiments

I started this blog for a number of reasons, one of which was to encourage myself to experiment and try new things.  I have always been curious about lapse videos.  I really have no excuse for not trying to make one earlier.  My cameras are equipped with an intravalometer  which is the necessary bit of camera equipment to make them.
 



 I knew how to setup the camera, but I didn't know what to do with the frames once I captured them.  A bit of googling led me to this page.   This was all I really needed.  However, there are tons of more complete resources out there to help also.  Here is a good overview.

I set my camera up to make a frame every 3 seconds.  I let it make 150 frames or so.  I had my camera set to create small JPEGS at basic settings.  Even dropping down to the small setting creates frames with more resolution greater than HD video.  I used the gimp to resize the frames to 720P resolution.  I let it change the aspect ratio on one video, and let it pad the video in the other.  I'm not sure which way I like better.  In the future I should probably compose with the aspect ratio in mind, and crop it down in post.  I created the videos with the following command:

>ffmpeg.exe -r5 -i DSC_%4d.jpg -an -vcodec copy drip.avi

It makes the video at 5 frames per second, making it about 30 seconds in length.  I didn't add any audio.  I like seeing how the shadows and trees move about in the videos.



Feel free to watch the videos, but remember these are the efforts of a beginner testing the water.  You will never get that time back.  You have been warned.

(Note: If you are viewing this in Reader or Buzz, the embedded objects don't seem to show up)

Sunday, March 22, 2009

Automation

We all find ourselves doing the same set of boring tasks to our images over and over again. Fortunately most of the tools we use to work on our images have ways to automate tasks.

In the Photoshop world it is called "Actions". In the Gimp world they are called "Script-fu". If you look at just the names, Script-fu wins hands down. You see the name "Script-fu", and you start to imagine yourself as some martial arts imaging master. Hiii-ya!

However, in practical use I imagine Photoshop wins. I do not have Photoshop, but this tutorial shows me how to creat an Action. It is as simple as "recording" what I do to an image! To get the job done with Gimp, you first roll up your sleeves. Then you learn some basic Scheme. Grab the documentation, some example code, and frustrate the heck out of yourself for hours.

I wanted to automate the smart sharpening technique I have used for most of my images. If I had to create this "Action" under Photoshop I imagine it might take me about 1 hour to learn how to record the tasks, and create Action that I like. However, I don't have Photoshop (yet), so I Script-Fu'd my Gimp. I now have a nice automated utility, but it took me about 3 hours to get there. If any wants my source, let me know I'll try and find a place to post it.

The images here weren't processed through my new script, but are the implied follow-ons to my last post!