I said FLV but meant SWF.
Main Topics
Browse All Topicshello all,
i have been awarded a project to assemble some high-res images which are presented within a Zoomify window.
at the moment, i have enlarged the zoomify window to fit my desktop (dual 1680*1050 = 3360*1050) and capturing parts, then assembling them with Photoshop CS4 PhotoMerge.
but with nearly 40-50 slices per image, and nearly 1000 images, it will take an eternity to stitch them all.
my zoomify window is set to 40000*40000 pixels in dimension so the whole image loads automatically without having to have a section visible(onscreen), for the data to load.
even adding two more displays to take me desktop size to 3360*2100 will not solve the basic problem.
can you suggest any automated alternative to my method??
a screen capture tool to grab an object, which extends beyond the screen (capturing 40000*40000 instead of 3360*1050) will prove to be the best solution.
looking forward to your thoughts.
hathehariken.
This Question has been solved and asker verified All Experts Exchange premium technology solutions are available to subscription members.
Experts Exchange has been collecting answers to technology questions since 1996…3 million and counting! If you have a question, chances are we already have your answer.
If you can't find the exact answer you're looking for, ask our exclusive community of 50,000 experts. You’ll get a personalized answer from a trusted professional.
Thousands of free tech tips, tricks, how-to’s and tutorials are available in our peer reviewed articles section. See for yourself how smart our experts are, no login required.
Access the answers to your technology questions today.
30-day free trial. Register in 60 seconds.
Members of the expert community talk about why the experience at Experts Exchange is different than what you will find anywhere else.

Try it out and discover for yourself.
30-day free trial. Register in 60 seconds.
Join the community of experts here and help other tech pros by answering question in your area of expertise. You can earn FREE access to all Experts Exchange's premium features and resources.
i do have accecss to the original 256*256 tiles - there are millions of them.
i tried assembling them, but since there is no overlap between adjascent tiles, the image assemblers are stumped (no reference data).
by backloading, if you mean background loading, i have already solved it by increasing the SWF <object> dimensions....
scrolling grabbers like Snagit, dont work.
the client lost the originals in a data disaster, so we are getting them back!
btw, I am in India :)
i found this program (written in Python, i think)
http://www.staremapy.cz/zo
what does it do?
thanks,
hathehariken.
Hi,
I am the author of the Python script mentioned above. Recently I was programming an image server compatible with Zoomify for serving dynamically from JPEG2000/TIFF images, I was also preparing compatible viewers in JavaScript so I think I can say I have expertise in image tiling. I am the person behind MapTiler.org application for Google Maps overlay generation.
It is not technically problematic to reconstruct the Zoomify tiles back to the original image, I just don't want to publish the program for that, because people will steal the online images and break the copyright - and I don't want to help them with that.
But in case you client has lost the originals and he is the owner of the image data then I can help you.
For stitching the tiles you can ignore the directories (TileGroupXX) and just get all the tiles to one directory. Every tiles has a form z_x_y.jpg, so you can process only the tiles with highest "z" to reconstruct the original image.
The tiles must be placed in the correct position of the grid with top-left origin.
I have a script for this, which runs well (for images which fits into the RAM of your computer).
In case you need to stitch images with size like 40.000 x 40.000 pixels then you need to handle it differently - as such image can be problematic even for Photoshop. You have to use a library like VIPS or GDAL to process such raster data and you must choose well the image format (correct version of TIFF, or wavelet based: JPEG2000, MrSID, ECW, ...).
Anyway, if you would like to have the merging script then contact me personally.
I can also create the modifications for the huge raster data stitching, or merge all your 1000 images from the web and publish somewhere the results (probably at Amazon S3) - but I will invoice the time I will spend on that.
Best regards
Klokan Petr Pridal
and the smiley was my (poor) attempt at humor!!
lots of work is offloaded to India - we are benifitting a lot from open markets.
on topic:
i dont know even a single drop of any snake, let alone Python.
so i think you will have to help me to set up the whole thing, if you can
i do have a local WAMP stack (Apace 2 Triad on W2kSP4), if a server environment helps in any way....
there is an order to the tiles - it is a 3d tiered stack collapsed to "n-x-y" nomenclature.
i spent 2 hours trying to visualise it, and all i got was a migraine!
but it is human readable, once you get the hang of it.
trying to directly manipulate the tiles turned out to be a disaster.
since there is no overlap between adjascent tiles, no image assembler / stitching software can handle these tiles in their raw form.
i tried some methods, and they all turned out garbage.
CS4 PhotoMerge, PTgui, Panavue - they all failed...
manually attaching the tiles to a grid will be a nightmare....
an idea!
every zoomify library comes with a XML file
-----<IMAGE_PROPERTIES WIDTH="2080" HEIGHT="3120" NUMTILES="169" NUMIMAGES="1" VERSION="1.8" TILESIZE="256" />
if a script can be made which can function like this:
-read the XML header which tells the total number of tiles and tile size
-calculate # of x-axis tiles and y-axis tiles
-create a table with # x and y cells.
-read the matrix and populate the table.
-display the table to the browser.
do you think this is possible with PHP+HTML??
from the browser, i can grab it with a scrolling grabber
i really appreciate you taking extra time to help me out.
regards,
hathehariken.
hello Klokan,
thank you for taking interest in this.
i already theorised that the highest Z is the 100% zoom level.
the pictures nowhere near 40k*40k pixels.
the maximum picture is around 15k*18k pixels.
my computer seems to handle these sizes fine, so i dont think i will require advanced processing algorithms. but i will keep this in mind.
i would like to have a look at your merging script.
i have mailed you at the address mentioned in your website.
hathehariken.
Don't know if you're still interested in answers for this, but I had the same problem and found a dead simple solution.
All I use is FlashGot in my Firefox browser and a built-in Photoshop feature called Contact Sheet.
You seem to have said you've got access to the individual tiles that Zoomify uses, so all you need to do is throw the address for them into FlashGot's 'Build Gallery' feature, select the image range for FlashGot to automatically display on the one screen. (Say the tiles are numbered from 4-0-0 to 4-10-8, then the range would be 4-[0-10]-[0-8])
FlashGot then brings them all up on a page for you, and then another click of FlashGot allows you to download them all with barely a click to whatever folder you specify.
From there you open up Photoshop's Contact Sheet feature (used normally to place smaller versions of images onto one sheet for professional purposes), select the image sizes and how many rows & columns of them there is. And voila. Done.
If you are still interested, I can go into better detail.
this photoshop contact sheet feature may be worth looking into.
but the real question is whether it can handle building a 50 mega pixel monstrosity...
i would be delighted to discuss more on this.
thanx,
hathehariken.
ps. the original assignment has been completed. i used PHP's GD and Python together to get the job done.
Business Accounts
Answer for Membership
by: D_BruggePosted on 2009-09-04 at 21:37:22ID: 25264966
That's a stumper.
I'm not an expert on backloading FLV files, but I don't think that the traditional methods would work. (You might want to post a question in the Flash zone, where the Flash Geeks hang out) I can't think of a way to bypass the Flash aspect since the new image isn't loaded until you scroll the window.
Can't think of a way to automate screen grabs either because it would be a considerable chore to integrate measured scrolling into a script, and be able to detect the image edges.
Is there any way of going to the source and asking for access to the originals?
Whenever I am confronted with a really tough project like this, I usually get someone in India to do it for me (wink)