Image XObject using /F file specification

We are having a problem where instead of the stream data being on the Image XObject itself, we want to use a stream that is an attached file (we are manually writing the pdf file at this point using internal tools). Inevitably, it keeps trying to get that file from the disk and failing.
This same file specification we are using as the /F property on the XObject dictionary is the same one in the as we have as an entry in the EmbeddedFiles name tree, and the file does appear appropriately in the attachments list. The file spec includes all the correct /F filename and /EF<</F reference>> to the actual data stream.
The XObject dictionary still has a zero byte stream associated with it, and the /Length property set to 0. And we're using raw RGB values, so no filter (/Filter or /FFilter) needs to be specified.
The error we keep getting is "This file cannot be found" (fileErrFNF). If we extract the file, then we get a different error of "You do not have access to this file" (fileErrPerm), unless supposedly, you are running Acrobat Pro, in which case it works as you would expect, but that doesn't solve the problem of us wanting to have it embedded as a file and be read internally, not from disk.
So the file specification should be correct, since we can see the embedded file in the attachments and extract it correctly. And the XObject dictionary should be set up correctly, since when the file is extracted, we can get it to render if running Pro. It's just that for some reason, we can't get the XObject to look for the EmbeddedFile instead of an external file.
We are using Acrobat Reader 9 for viewing and building this against the "PDF Reference, version 1.7" specifications. This also fails when using AR8 as well.
Options that we have tried varying are not including the EmbeddedFiles name tree entry (thus making it invisible), including or excluding the UF or DOS/Mac/Unix values on the file specification dictionary and /EF dictionary. We've tried removing the base /F value entirely, and only had the /EF dictionary. What else can we try to get this to work?

Here are the objects as have I described above:
63 0 obj
<<
    /Type /Filespec
    /F (pic1.rawrgb)
    /UF (p i c 1 . r a w r g b )
    /Desc (embedded image)
    /EF
    <<
        /F 125 0 R
        /UF 125 0 R
    >>
>>
70 0 obj
<<
    /Names [(pic1.rawrgb) 63 0 R]
>>
99 0 obj
<<
    /BitsPerComponent 8
    /ColorSpace /DeviceRGB
    /Width 200
    /Height 200
    /Subtype /Image
    /Type /XObject
    /F 63 0 R
    /Length 0
>>
stream
endstream
125 0 obj
<<
    /Type /EmbeddedFile
    /Length 120000
>>
stream
% 120000 bytes of 8 bits per color RGB for a 200x200 pixel image without any filtering
endstream

Similar Messages

Image-to-Image Fade Using External Files

Hi all,
I'm trying to create a fairly basic random image-to-image fader. I've got a LOT of images that need to rotate through and they've all been optimized, however the swf file just becomes too big w/all of them.
So what I was hoping I could do was use AS3 to load up a random jpg from an XML file, apply the fade effect, preload the next one it's planning on grabbing while the current one is being displayed, and repeat the process indefinitely. By doing this I'm hoping to keep the footprint of the application extremely small.
Is such a thing possible (I'm using a trial of CS5 at the moment) and if so, can someone point me in the right direction? I'm pretty new to AS3 so any amount of help would be greatly appreciated!

Yes, it is possible to do. You can probably search Google using terms like "AS3 XML slideshow tutorial" to get the whole thing in one package. Though what you might want to do is take it in steps so that you can come to understand how each aspect works.
The basic elements would involve loading an XML formatted bit of data into an array and using the Loader class to load the images, using a COMPLETE event listener so that you know when an image is loaded and ready to be processed into an object within the show.
Then you would want to look into using some actionscript tweening to get the alpha fading effect working for you. Flash has its own Tween class, and there are thrid party tween classes as well. I'd say use the Flash version first just to eliminate one level of complication for now (installing and using an external class). Many people recommend the thrid party tweening classes over Flash's.
You might also want to incorporate some form of Timer (the Timer class) to help manage when things change and/or when they load. If you have optimized your images already, then it would take less time for them to load, which means you probably do not want to use loading status to control the timing, but only as logic to enable timed actions.

Masters in Aperture Library Using Referenced Files

I use a MBP with an SSD to hold my Aperture 3 library.
I use a G-RAID to store my images and use Referenced Files.
All was good until yesterday, I deleted my Thumbnails to recreate them and remove the orphans.
After restarting Aperture and the processing time to recreate the Thumbnails (AP.Minis, Ap.Thumbnails, and AP.Tinies) I noticed my AP library was huge.
After inspecting the contents, there is now 65Gb worth of Masters, why would these be there since I use Referenced files and why would they have just now shown up after rebuilding my Thumbnails?

Remember:
-- A RAW file cannot be edited. It can only be written in a camera (or scanner, etc.) To be seen, it must be read through a RAW developer. The RAW developer (or converter, if you prefer) can apply changes to the bitstream (called "adjustments" in Aperture) but those changes always live in a separate file, called a "version" in Aperture. EVERY master image file in Aperture has at least one version file. The two are combined, in real time, every time you see a file. Thus questions about the Aperture library becoming corrupted are a bit moot - if you lose your library you will lose your edits. The only way to "save" edits to a RAW file is either adjust it in Aperture (or Lightroom) and let the program combine the two, or convert the file to an industry standard format, like TIFF or JPEG.
-- Aperture uses a "non-destructive" workflow. This means that it never really edits ANY file - not even a JPEG. This avoids the problem of recompressing JPEGS each time they are opened and close. So even JPEGS remain unaltered with your "edits" or "adjustments" being stored in one or more versions.
The only time you need to "save" a file in Aperture is when you need a new file in a specific format to give to someone. Until then, Aperture happily keeps track of your original images (managed or referenced) and the corresponding "edits" which it stores in a version linked to the original image file.
There are many reasons to prefer managed over referenced or vice versa, but fears of database integrity should not be one of them. You are making a lot of work for yourself - put the Aperture Vault on your server and/or use Time Machine. "Exporting" is a terrible way to try to make backups - it will take forever and you will end up with files scattered all over the place, defeating the purpose for which you bought Aperture in the first place.
Hope this helps.
DiploStrat

XObject - How to use File specification dictionary

Hi all,
b Here it is a complete PDF file with just a small picture in the page and it works well (windows XP Professionel SP 2 and Mac 10.3.9)
%PDF-1.7
1 0 obj [/PDF] endobj
4 0 obj << /Length 36>>
stream
q 100 0 0 100 100 400 cm /Im3 Do Q
endstream endobj
5 0 obj <</Type /Page /Parent 2 0 R /MediaBox [0 0 595 842] /Contents [4 0 R ] /Resources <</ProcSet 1 0 R /Font <<>> /XObject <</Im3 3 0 R >> >> >> endobj
3 0 obj <</Type /XObject /Subtype /Image /Width 39 /Height 41 /ColorSpace /DeviceRGB /BitsPerComponent 8 /Filter /DCTDecode /Length 1087>>
stream
% binary data
endstream endobj
2 0 obj <</Type /Pages /Kids [5 0 R ] /Count 1>> endobj
6 0 obj <</Type /Catalog /Pages 2 0 R >> endobj
xref
0 7
0000000000 65535 f
0000000010 00000 n
0000001536 00000 n
0000000278 00000 n
0000000033 00000 n
0000000120 00000 n
0000001593 00000 n
trailer
<</Size 7 /Root 6 0 R >> startxref 1642
%%EOF
b Now, I add a file specification for C:\Epinette.jpg (or /Epinette.jpg at the root of HD Disk on Mac) using Table 3.40 and the example page 348 (alternates) and I get a blank page. It does not matter if I let the stream or I write an empty stream (normally the viewer ignores the stream ; so its correct on this point). The file exists and is readable by Photoshop. I put in the same directory or folder (so root C:\ or root /) the PDF file. I am connected as Administrator.
b Here it is the second example.
%PDF-1.7
1 0 obj [/PDF] endobj
4 0 obj << /Length 36>>
stream
q 100 0 0 100 100 400 cm /Im3 Do Q
endstream endobj
5 0 obj <</Type /Page /Parent 2 0 R /MediaBox [0 0 595 842] /Contents [4 0 R ] /Resources <</ProcSet 1 0 R /Font <<>> /XObject <</Im3 3 0 R >> >> >> endobj
3 0 obj <</Type /XObject /Subtype /Image /Width 39 /Height 41 /ColorSpace /DeviceRGB /BitsPerComponent 8 /FFilter /DCTDecode /Length 0 /F << /Type /Filespec /F (/C/Epinette.jpg) /UF (˛ˇ/C/Epinette.jpg) >> >>
stream
endstream endobj
2 0 obj <</Type /Pages /Kids [5 0 R ] /Count 1>> endobj
6 0 obj <</Type /Catalog /Pages 2 0 R >> endobj
xref
0 7
0000000000 65535 f
0000000010 00000 n
0000000516 00000 n
0000000278 00000 n
0000000033 00000 n
0000000120 00000 n
0000000573 00000 n
trailer
<</Size 7 /Root 6 0 R >> startxref 622
%%EOF
b The two characters before the path in UF parameter are thorn and ydieresis. I tried to delete C in the path (no drive). The same in PDF 1.4 (without UF parameter) on windows or Mac. I tried with and without /Mac in the text string as the table 3.40 shows. Idem for HD.
Sure, I will send the two PDF files for someone if necessary.
Thanks for help.
François

Thank you for the very interesting article written by Leonard Rosenthol.
It is a good and clear answer for this problem.
But it is pity.
I use to Reader on PC and full Acrobat Professional 7.0.9 on Mac.
Thanks
François Faguer

We are currently looking for a way to link images to a design file within programs like InDesign and Illustrator using an HTML link instead of a local file. We are hosting our images in SharePoint and need the design file to retain it's links, no matter

We are currently looking for a way to link images to a design file within programs like InDesign and Illustrator using an HTML link instead of a local file. We are hosting our images in SharePoint and need the design file to retain it's links, no matter who on our network opens the design file.

The Cloud forum is not about using individual programs
The Cloud forum is about the Cloud as a delivery & install process
If you will start at the Forums Index https://forums.adobe.com/welcome
You will be able to select a forum for the specific Adobe product(s) you use
Click the "down arrow" symbol on the right (where it says All communities) to open the drop down list and scroll

When I have video clips on my timeline, they do not play in the monitor screen. I hear the sound track but can not see the image. To test if my new files are somehow the problem, I went back and used video files that I used successfully a year ago, and th

When I have video clips on my timeline, they do not play in the monitor screen. I hear the sound track but can not see the image. To test if my new files are somehow the problem, I went back and used video files that I used successfully a year ago, and they no longer play in the monitor window. I hear the sound track, but do not see any video. Have I changed some setting that controls video playback in the monitor window?

Randy Ruttger
Thanks for the follow up.
Missing in this equation were the version of Premiere Elements that you were using and the computer operating system on which it was running. But...
The deed is done. But.....Until now and after you went ahead and moved to version 12, we now learn what version of Premiere Elements you were using. We still do not know the computer operating system involved.
You just said Premiere Elements 10. Premiere Elements 10 is hit by a serious known display issue if the computer uses a NVIDIA GeForce video card/graphics card. In this issue, the only remedy is to roll back the driver version to about May 2013. The description of the problem and how to do the roll back are described in one of the Announcements at the top of this forum. What video card/graphics card does your computer use?
On another front, Adobe will be releasing a new version of Premiere Elements any day now. No one but Adobe knows the feature set for the new version. That type of information is announced at the moment of release of the new version.
So the remedy to the problem that you presented in this thread could have been...
1. Rolling back the NVIDIA GeForce driver version to about May 2013 if possible (Windows 8 or 8.1 64 bit, may not be possible)
2. Move to another version of Premiere Elements (which you did)
I offer the above for consideration so that you can review your decisions. We are pleased to learn that Premiere Elements 12/12.1 works for you.
Please do not hesitate to ask questions and ask for clarifications, but remember that the answers are in the details.
Best wishes,
ATR

In CS4 Use of File Place Does Not Center Images

I have been using CS3 for about 1 1/2 years, so of course had to give CS4 a try. I downloaded the trial. I immediately had an issue with File > Place.
I often use File > Place to add additional items into a document. I believe the expected functionality of File>Place is that the file should then be centered in the document.
Here is what I have done:
+ Create a 12"x12" document at 200ppi (I have also tried to create a document at 2400x2400 pixels)
+ Use File>Place to add a square 2400x2400 pixel image (JPG file) into the document
+ The added layer is now offset ever so slightly in the new document.
+ Further observation shows that the horizontal reference point is at 1202 (screen shot attached), thus creating an offset from center. I change it to 1200 and sure enough the paper is centered perfectly. I thought maybe changing it once would make it the same next time, but that is not the case.
+ When I used the File>Place command, I was zoomed out (25%), and the entire image was viewable on screen. If I zoom out further, say 8.33%, the offset is 1206.
I am able to duplicate this on two different computers with the following configurations:
Computer #1
+ Windows XP Professional Version 2002 SP2
+ AMD Athlon 64 X2 Dual Core Processor 6000+ (3.01GHz)
+ 4GB RAM
+ nVidia GeForce 8600 GT, driver version 6.14.11.5819
Computer #2
+ HP Pavilion dv9843cl Laptop
+ Windows Vista x64 SP1
+ Intel Core Duo processor T8100 (2.1 GHz)
+ 4GB RAM
+ nVidia GeForce 8600M GS with 512MB DDR dedicated graphics memory, ForceWare version 167.43
+ Video BIOS version: 60.83.4f.00.22
Operating System: Windows XP Professional
Browser: Firefox
Video display card / driver version: nVidia GeForce 8600 GT, driver version 6.14.11.5819
Screen shot (not sure if this works in these forums):
[IMG]http://cdn-1-service.phanfare.com/images/external/1466784_1419796_53071200_Full_3/0_0 _41325929f3d5a1780af10559af98d829_1[/IMG]
Any ideas?????
~ Kathleen

"Place" ..... It is not all the easy to understand how place works without studying its behavior. Often it will work in a way your not expecting if you do not understand how it works. There is a setting in Photoshop preferences Resize Image during Place. Un Checking this setting does not mean the image will not be resize nor does checking this setting mean that it will be resize. On top on top of that the seems to be something in Photoshop on My system that keeps changing the setting to the checked state.
What the heck am I writing am I trying to make this worse. No I'm not. To prove that I first tell you it is very easy to center position a placed layer in an action after the Place step add a step select all the two layer>Align layers to selection>Horizontal Center then vertical center. The deselect...
Place....
There is a lot going on remember in Photoshop a Layer can be any size Larger/smaler/same as the canvas and can have any aspect ratio. If the placed image has more pixels them a canvas side size and the preference is checked the placed image will be transformed to fit within the canvas size. If not checked the placed image will transformed to the same dpi resolution so no pixels will be changed. The only time a place image is not transformed is when the the preference is not checked and the placed image file has the same dpi as the current Photoshop Document. In this cast the image is placed in at 100% size be it smaller/larger/same as canvas size.

Why do the file extensions (.jpg .gif .png) no longer appear when I click on a previously saved image to use that image's file name (particularly important when saving a series of images using the same root name)?

I save a lot of images using firefox, often times from a large batch or series of images. It used to be that I would click on a previously saved image and the entire file name including the file extension (i.e. image_example.jpg) would appear in the "save as" line. Now when I click on a previously saved file, the file name appears without the file extension (i.e. image_example). Which means I have to manually type .jpg every time. For a large collection of images that I am hoping to use the same root file name and then add chronological numbers at the end, this has become incredibly frustrating, especially as it is a previously unnecessary task.
I am using a new Macbook Pro and maybe there's something Apple related to this...? It did not happen on my old PowerBook G4. I have file extensions turned on in System Preferences.
It should be noted that I have searched high and low and have even gone into the Apple Genius Bar where they were just confused as I was and of course ended by urging me to use Safari (shocker!) as it has all kinds of new extensions and bells and whistles. I seriously feel alone on an island with this dumb, hard to google problem. Thanks so much for any help anyone out there might have.
I mean: is this as simple as changing a setting in about:config?
Your assistance is greatly appreciated.

Thanks for your response Mylenium, however like I mentioned multiple times, I did change all of my trackpad/scrolling settings in system preferences. And if I wanted to use a normal mouse (or a tablet), I would've gotten an iMac instead of a MacBook Pro. I travel often and work all over the place, not always with access to a decently sized workspace that would be required for using a mouse or tablet.

[svn] 1978: Bug: vendors. properties file which is used in vendor specific login commands was not being read properly and as a result some login related error messages were not being displayed correctly .

Revision: 1978
Author: [email protected]
Date: 2008-06-06 08:05:34 -0700 (Fri, 06 Jun 2008)
Log Message:
Bug: vendors.properties file which is used in vendor specific login commands was not being read properly and as a result some login related error messages were not being displayed correctly.
QA: Yes - we need automated tests to make sure that errors.properties and vendors.properties in BlazeDS/LCDS are loaded properly.
Doc: No
Modified Paths:
blazeds/branches/3.0.x/modules/common/src/java/flex/messaging/util/PropertyStringResource Loader.java
blazeds/branches/3.0.x/modules/opt/src/jrun/flex/messaging/security/JRunLoginCommand.java
blazeds/branches/3.0.x/modules/opt/src/tomcat/flex/messaging/security/TomcatLoginCommand. java

I have a lot of grief with this version of Windows Media Player.
It is very buggy and frustrating to use.
I have my Music library on a QNAP NAS, which is as reliable as they come.
System notifications make it not save changes. It also does not do a good job of interpreting albums and artists from folders. Changes to track names are not saved, nor are tracks moved to other albums, renamed albums, changes to genre, artist
or date. It separates and merges albums/tracks without sense or reason. Some changes I've made up to 4 times, then closed WMP and re-started my machine to check if it has/hasn't saved the changes. Often it has not.
This is the first time I've used WMP in this capacity, and I do not recommend it.
New service pack please.

Using Jar file images with READ_IMAGE_FILE built-in

Hello,
I was wondering if it is possible to utilize images in a JAR file with the READ_IMAGE_FILE built-in. The Oracle Forms Configuration guide seems to indicate that I should be able to do this by using 'URL' with READ_IMAGE_FILE but no success so far. I have imageBase=CodeBase in the FrmsWeb.cfg file and the jar file in the /forms/java directory. Any advice would be most welcome,
regards,
Kevin.

It is possible.
in 10g 'URL' is a new value for file_type in READ_IMAGE_FILE.
It's not documented in Forms Help, but should work.
1. Check Java Console and ensure jar file is loaded
You should see something:
Downloading http://host:port/forms/java/myjar.jar to JAR cache
or
Loading http://host:port/forms/java/myjar.jar from JAR cache
2. You are using dept for file name. Check the image file name in your jar file. It should be full, case sensitive name of JPG or GIF file in your jar.
like :
READ_IMAGE_FILE('dept.jpg','URL','control.deptimage');
3. If all above correct, try to put synchronize; before READ_IMAGE_FILE

Using a file as a source of images for a slideshow in CS6

I was wondering if someone might be able to help me create a slideshow that uses a file to pull images from rather than calling on an individual image source code. As it is right now, I have a basic working jquery slideshow. What I am wanting to get to though is instead of having the numerous <img source=..../> code I could instead have the slideshow automatically pull images out of a designated file. Then, to add images, I could just import the pictures into the designated file. I am very new to using Dreamweaver and any sort of coding in general, so I do not even know if this is possible or worthwhile. Any help will be greatly appreciated!

You can't pull files from files. But you can certainly pull files from a designated folder or directory on your server. This example below uses PHP code with the jQuery Fancybox Viewer.
Alt-Web :: PHP & Fancybox Photo Gallery
Nancy O.

Centre aligning images within picture frame using XML file

Hi
I am using a flash file that uses an xml file to upload a
series of images one at a time into a picture holder.
There are two different widths of the images. The height is
always the same. Is it possible to set this up so every image that
the xml file calls is centre aligned within the picture frame.
At the moment when using the xml file to load the image it
always left aligns the image when it brings it into the picture
frame even if you change the alignment to centre align within in
the align panel.
Does anyone have a suggestion why this is happening or how to
fix it?
Thanks for the help

You're welcome. Pleased it worked.
All it does is use the MovieClipLoader class to load the
images into your holder_mc. Because loading takes time, and because
you can't check or adjust the size or position of images until they
have loaded, the MovieClipLoader class gives you a handy option to
run some code whenever its finished loading something you told it
to load.
That's what the addListener instruction does... it says
whenever myLoader has something new to tell me about the status of
the load requests I ask it to do... make it run some related code
in the listener object (which is called loadListener in this
code... it could equally be called something else). You 'listen'
for different events or updates of status. All I 'listened' for
here was the onLoadInit event.
The onLoadInit is the 'final' status update after loading. It
occurs when something has loaded and its properties are accessible
(_x, _width etc). The target_mc that is an argument for the
loadListener.onLoadInit is a reference to the clip that was loaded
into.
The best way to learn about this stuff is probably to read up
on MovieClipLoader in the livedocs and try google to see if there's
an online tutorial somewhere. Then give it a try yourself in
another new fla.
cheers
GWD

Encountering ORA-39000: bad dump file specification while using datapump

Hello,
I am trying to use datapump to take a export of a schema(Meta_data only). However, I want the dump file to be named after the date & time of the export taken.
When i use the following command -- the job runs perfectly.
expdp system@***** dumpfile=expdp-`date '+%d%m%Y_%H%M%S'`.dmp directory=EXP_DP logfile=expdp-`date '+%d%m%Y_%H%M%S'`.log SCHEMAS=MARTMGR CONTENT=METADATA_ONLY
However, I want to run the export using an parfile. but if use the below parfile, i am encountering the following errors.
USERID=system@*****
DIRECTORY=EXP_DP
SCHEMAS=TEST
dumpfile=expdp-`date '+%d%m%Y_%H%M%S'`
LOGFILE=MARTMGR.log
CONTENT=METADATA_ONLY
expdp parfile=martmgr.par
Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.2.0 - 64bit Production
With the Partitioning option
ORA-39001: invalid argument value
ORA-39000: bad dump file specification
ORA-39157: error appending extension to file "expdp-`date '+%d%m%Y_%H%M%S'`"
ORA-07225: sldext: translation error, unable to expand file name.
Additional information: 7217
How do i append date&time to the dumpfile name when using a parfile.
Thanks
Rohit

I got the below error while using the dumpfile parameter as dumpfile=dump_$export_date.dmp
Export: Release 11.2.0.2.0 - Production on Thu Feb 7 16:46:22 2013
Copyright (c) 1982, 2009, Oracle and/or its affiliates. All rights reserved.
Connected to: Oracle Database 11g Enterprise Edition Release 11.2.0.2.0 - 64bit Production
With the Partitioning option
ORA-39001: invalid argument value
ORA-39000: bad dump file specification
ORA-39157: error appending extension to file "dump_$export_date.dmp"
ORA-07225: sldext: translation error, unable to expand file name.
Additional information: 7217
Script i used is as follows
export ORACLE_HOME=/orcl01/app/oracle/product/11.2.0.2
export ORACLE_SID=$1
export PATH=$PATH:$ORACLE_HOME/bin
echo $ORACLE_HOME
export export_date=`date '+%d%m%Y_%H%M%S'`
echo $export_date
expdp parfile=$2
parfile is
DIRECTORY=EXP_DP
SCHEMAS=MARTMGR
dumpfile=dump_$export_date.dmp
LOGFILE=MARTMGR.log
CONTENT=METADATA_ONLY

Using swf file for very large static image

i have a large photo that i need to use as a banner image in a website. i have tried optimizing it for the web in fireworks -- to reduce the file size obviously -- but the client isn't happy with the resolution. i haven't had this issue come up before but i see his point (to a degree), as the detail and sharpness of the original is far superior.
i have almost no familiarity with creating with flash, but i remember reading somewhere that swf files do not have the same sizes as jpeg images.
would it be feasible to include the single static image as a swf file to preserve its original resolution? i mean, would that image as a swf file be smaller than the image as a jpeg?
mille grazie.

There are two things at play here... image quality (often harmed by compression) and image resolution (# of pixels width by height). What is the real problem?
Whether you render your image as a GIF, JPG, PNG or SWF file, they will all have the same resolution.
Packing an image in SWF should be a last resort as not everyone has Flash installed. For example, iPhone users cannot see SWF.

Figuring out how to extract images from a PDF file

Hi,
I'm trying to write a small app that extracts all images from a PDF file. I already wrote a nice parser and it works good but the only problem is that I can't quite figure out from the reference how to decode all images in a PDF file to normal files such as tiffs, jpegs, bmps etc. For now I'm focusing on XObject images and not dealing with inline images.
From what I understand so far just by trying and looking at open sources I figured that if I see a XObject Image with a DCTDecode filter, taking the stream data without doing anything to it and saving it as a jpeg file works. But doing the same to FlateDecoded streams or CCITTFax didn't work.
What is the right way to properly extract the images?

In general you have to
* decode the stream
* extract the pixel data
* use ColorSpace, BitsPerComponent, Decode and Width to unpack pixel
values
* reconstruct an image file format according to its specification
There are no other shortcuts. The DCTDecode shortcut (which doesn't
work for CMYK JPEG files) is just a piece of fantastic good luck.
Aandi Inston

Image XObject using /F file specification

Similar Messages

Maybe you are looking for