Speeding up image analysis,

I am using a VI to analyse an AVI file showing liquid particle flow
I want to acheive this in real time, by using the video capture card, however the system is far too slow to acheive this
I am using colour matching, and doing some binary operations to identify particles from their background.
im using a pentium 4 3ghz, with gig of ram, but this still runs slow
how can i speed up the vi?
also, i would like to use an overlay function to show the particles in colour, by removing the background using binary operations, i.e. seperating the particles from their background in binary, then retriving the colour image and only displaying pixels that are set to logic one

What are the specs of your video capture card? Could it be too slow for realtime acquisition and transfer to memory and thus the bottleneck?
Is color needed, can you use colored filters, enhanced lighting, or use narrow band light sources to increase the S/N ratio so software has to do less processing?
Can you do an ROI so that you do not need to process the entire image frame?
Do you have enough pixels/particle for sufficient resolution.
Are you using NI Vision Builder? If so, you can benchmark the execution time of each individual vision function. It is an awesome package to do vision prototyping, fooling around, and generating LV code.
Machine vision is a science and an art with a good measure of experience and luck thrown in.
~~~~~~~~~~~~~~~~~~~~~~~~~~
"It’s the questions that drive us.”
~~~~~~~~~~~~~~~~~~~~~~~~~~

Similar Messages

Speeding up image icons.

Is there any way to cache or speed up image icons in the Finder? For example, I have a folder full of huge .HDR images. It takes several minutes for the entire folder to build the icons for all the files.

Rather than post new question, I'll bump this one.
Also trying to figure out if there is a way to cache the preview/quicklook/coverflow/whatever icons for Finder.
The lag time everytime a folder is opened up with a lot of items is really annoying.
Thanks.

Image analysis with PS CS3 EE

First off, I need to apologize if I am not very clear on this issue Im fairly math challenged and know enough about Excel to fill out a time sheet and make an invoice. If this kind of thing is up your alley, you are just what were looking for.
We are attempting to use the image analysis tools in CS3 EE and Microsoft Excel to determine a number of things in order for a set of photo images to meet a given set of requirements. Most of this has to do with examining histograms and determining things like average pixel levels for each of RGB channels, % of pixels in a given range, and percentage of the image that is clipped. This analysis would be done on a random sample of files in order to show the client that the images meet their standards.
Some starting questions:
From what I see of the CS3 analysis tool, it seems to zero in on the gray channel information (perhaps this was developed for the medical diagnostic community) and Im wondering if you can export the same information (Gray Value Minimum-Maximum-Mean) for the R, G and B channels.
Does anyone already have the Excel formulas for extracting this information for the Histogram.csv that is generated by the analysis features?
This ones a bit murky - is there any method to calculate the numeric value of Contrast (the difference between the histogram reading at .01 and 99% and have that as exportable data?
This one is more like a feature request (or is there another plug-in or other software to do this):
Is it possible to view all three channels (RGB) with the same detailed histogram information as the main view, without having to toggle between the various bands?
Anyway, I know this long and I appreciate your time. I will follow this post closely and if you think you can help Id be happy to go off forum with this.
Thanks! TLL

I'm envisioning doing this in AppleScript, and am assuming a histogram in JS is the same...
>% of pixels in a given range
Divide 256 by however many "range segments" you need. For simplicity, let's say 3: highlight, midtone, and shadow ranges. So (roughly) levels 0-85 are shadows, 86-171 are midtones, and 172-255 are highlights. You would iterate through each range of the histogram counting the number of pixels for that range. Divide each count by the total number of pixels.
>percentage of the image that is clipped.
The amount of clipping can be determined by how many pixels are dedicated to levels 0 and 255 of the histogram. If those levels have a high number of pixels, then you have clipping. You have to define for yourself what "high number" means, but in a "perfect" photo, you won't have pixels at levels 0 or 255 (pure white and pure black).
Carl.

Image analysis

Hi, I would like to know if the Cisco IronPort email security supports the Image analysis, If yes what is there any license required? Thank you

Yes, its under "Security Services>Image Analysis"
Its does require a feature key.
Check with your reseller for pricing... they might be able to get you a demo key so you can see if it catches what you want it to catch...

Intermedia Image Analysis cant find DLL

Win2k, Oracle 8.1.6EE. Fresh installations.
With a new installation of Oracle, create a SHARED database using connection pooling etc. On another identical machine also create a standard default database from the CD.
On both the standard default database and the shared one, can set the properties of images in ORDVIR objects.
Only on the default can I analyse the images. On the shared I get the following error:
err:ORA-06520: PL/SQL: Error loading external library
ORA-06522: Unable to load
DLL at 10000100
I've not changed the SQLNET, TNSNAMES or LISTENER since the setup did it for me as it looks OK, and as I say - works for setting the properties, just not the image analysis.
Any ideas?
Here are the files:
# SQLNET.ORA Network Configuration File: C:\Oracle\Ora81\network\admin\sqlnet.ora
# Generated by Oracle configuration tools.
NAMES.DEFAULT_DOMAIN = web.cre8tiv.com
SQLNET.AUTHENTICATION_SERVICES= (NTS)
NAMES.DIRECTORY_PATH= (TNSNAMES)
# LISTENER.ORA Network Configuration File: C:\Oracle\Ora81\network\admin\listener.ora
# Generated by Oracle configuration tools.
LIVELISTENER1 =
(DESCRIPTION_LIST =
(DESCRIPTION =
(ADDRESS_LIST =
(ADDRESS = (PROTOCOL = TCP)(HOST = rioja)(PORT = 1521))
(ADDRESS_LIST =
(ADDRESS = (PROTOCOL = IPC)(KEY = EXTPROC0))
SID_LIST_LIVELISTENER1 =
(SID_LIST =
(SID_DESC =
(SID_NAME = PLSExtProc)
(ORACLE_HOME = C:\Oracle\Ora81)
(PROGRAM = extproc)
(SID_DESC =
(GLOBAL_DBNAME = islive1.web.cre8tiv.com)
(ORACLE_HOME = C:\Oracle\Ora81)
(SID_NAME = islive1)
# TNSNAMES.ORA Network Configuration File: C:\Oracle\Ora81\network\admin\tnsnames.ora
# Generated by Oracle configuration tools.
ISLIVE1.WEB.CRE8TIV.COM =
(DESCRIPTION =
(ADDRESS_LIST =
(ADDRESS = (PROTOCOL = TCP)(HOST = rioja)(PORT = 1521))
(CONNECT_DATA =
(SERVICE_NAME = islive1.web.cre8tiv.com)
(INSTANCE_NAME = islive1)
EXTPROC_CONNECTION_DATA.WEB.CRE8TIV.COM =
(DESCRIPTION =
(ADDRESS_LIST =
(ADDRESS = (PROTOCOL = IPC)(KEY = EXTPROC0))
(CONNECT_DATA =
(SID = PLSExtProc)
(PRESENTATION = RO)
Thanks, Jason.

Thanks - this does appear to be the problem.
I can reproduce this affect on any clean machine with a fresh install so there is obviously an Oracle bug here.
Running the command
select FILE_SPEC from all_libraries where library_name = 'ORDVIRLIBS';
gives the output
C:\Oracle\Ora81\lib\libordvir8.so
for any non-default installation of a database created using the DBCA which uses InterMedia&Virage.
The file libordvir8.so doesn't exist.
It should have been oraordvir8.dll in the bin directory which does exist.
Thanks,
Jason.
<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by lguros():
You may want to do a
select library_name from all_libraries where library_name = 'ORDIMLIBS';
and
select library_name from all_libraries where library_name = 'ORDVIRLIBS';
and make sure the FILE_SPEC field is valid and points to a correct shared object or DLL.
If you DB location was moved, these will no longer be valid.<HR></BLOCKQUOTE>
null

GUI Tool for Image Analysis (delimited ascii file)

i've created a Java GUI Tool for Image Analysis. It does length calculation and locate the coordinate of centroid of an image...the problem now is how do i save the results in a delimited ascii file??

Create a FileOutputStream, use that to construct a PrintWriter, then use the PrintWriter as you use System.out. Print out the data, then append the delimiter manually after each set of data.
To get the file, the simplest ways would be to hard code an output file name or at least have a text field so the user can enter it. If you want it to look more professional, use a JFileChooser.
I hope this helps!
-JBoeing

Help with image analysis Photoshop CS4 Extended

Hi there - I recently went to the Birmingham NEC photo expo and saw a guy there giving a CS4 extended demo. He showed how at the click of a button a folder of new photos were reduced to a smaller bunch of just the most tack-sharp ones. I can't find out how to do this with my CS4 Extended - can anyone help - or am I just imagining it all :-)))

Sounds interesting.
Did he do that in Photoshop itself or Bridge or Lightroom?
Anyway, I suspect one could maybe achieve something like this in Photoshop with some fairly heavy Scripting, but I suspect the Script would have to actually open the images for analysis …

Urgent ! Speed and image size problem.

Dear Java Guru,
I have the speed caused by:
param.setCompression(TIFFEncodeParam.COMPRESSION_JPEG_TTN2);
If use:
param.setCompression(TIFFEncodeParam.COMPRESSION_DEFLATE);
It will be 6 times faster, but the image size is bigger.
But we need to keep the image size very small, say the COMPRESSION_JPEG_TTN2
allows the size to be 2714kb for 20 pages (Grayscale at 100dpi resolution). Is there a compression method that allows the image size to be 1500kb for 20 pages, yet at the time of encode, it is as fast as _DEFLATE method ?
Please help !!! our case is urgent
Thanks in advance.
Thinkwork.

Please help !!! our case is urgentIt always is, isn't it?

Question on image analysis.

Hi,
I looked into tutorials on Java2D which teaches me on transforming images.
But I want to analyse images (in common formats like JPEG). For example, I
want to get, say, % of area in yellow color, % in red, green etc. Something
like that. I presume this would require to read the image line by line,
pixel by pixel and get the attributes (color) of each pixel. Is it the
correct direction for me to study tutorials on iostream (to read the image
file)?
Any guidelines and reference for me?
thanks in advance,
jackling.

I haven't kept up with the new fangled java image stuff, but the way I used to play with pixels was as follows:
pixelArray = new int[width * height];
try{
       pixelGrab = new PixelGrabber(rawImage,0,0,width,height,pixelArray,0,width);
            if(pixelGrab.grabPixels() && ((pixelGrab.getStatus() & ImageObserver.ALLBITS) != 0)){
            } else {
                //throw some kind of exception
        } catch(InterruptedException e) {
            System.out.println("MapInteraction::ripImage ~ failed to grab pixels from image");
    }then you can do manipulations using the ARGB channels for each individual pixel. The array that is generated holds scan lines of the original image in sequence, so therefore pixel 1,1 (from top left, starting @ 0,0) would be @ pixelArray[width*1 + 1].
Example of bit analysis using bitwise operators:
if ( (pixelArray[y*width + x] & 0x00FF0000) == 0x00FF0000 ) {
Make sure your in 32-bit colour resolution when you do the ripping. I've had issues in the past when I was doing analysis expecting 32-bit mode and forgetting to set it!!

I can't use the Image analysis tools

Whenever I go to Image and then Analysis all the options are grayed out except the ruler tool. I would like to set a custom measurement scale and record measurements but they're grayed out. Please can someone help.

Did you check for an update of that "image toolbar' extension? <br />
Have you tried to contact the developer of that toolbar to him know there's a problem with '''his''' extension?

Very slow buffering speed & webpage / image loadin...

Hello,
about a month ago this whole problem started and i've tried almost about everything i could find on the internet to fix it, trying different versions of flash.. java different web browsers reinstalling my OS and buying a new BT Hub, but nothing has fixed it.
My download and upload speeds are what they were and have not changed at all, but when i view images, GIFs, youtube videos ect everything loads very slowly and i even get forced to watch youtube videos in 240-320p - before this happened i could watch them in 1080p and they would buffer in a few seconds.
The image and GIF loading is basically like im on Dial-Up again and portions of the image load every few seconds, also the same for netflix buffering and spotify song buffering.
Any help would be greatly apprichated because this is very frustrating.

I was just checking that there wasn't a problem with the graphics/video card. Are you connecting to the Internet by wire or wireless?
Have you checked that the Network card drivers are up to date. Use the card manufacturer's website rather than Windows which can be out of date.

Speed up image / page loading

I've done a remake of the website for my company. I'm a hobbyist web person - not my full time job here. My boss wanted lots of images. Well - now that we have them, he says the page takes too long to load. Any pointers on how to get the images / page loaded faster? The template I have based the site on has an ajax pre-loader but it doesn't seem to actually pre-load all the images. The galleries still take a while to load when you view them. The site is up at http://www.morgansmithllc.com. Any suggestions on how to speed things up that a hobbyist user of DW can implement? Thanks!

This is one of the downfalls of a single page layout. In that type of layout, everything must be loaded because it all exists on the single page.
As a result I tried running your page through some analyzers and found there are some extremely large files that are being used. The backgrounds are huge and could likely be further compressed. Right now your top 6 images are 10MB of the 20+MB page size which is far too large. Also I see a lot of parallax javascript hanging around. Just because a website is a single page does not mean it needs to be a parallax site. In fact, for this type of site I would recommend removing the parallax because there is no need for it.
If you simplify the site in those ways, that should help a good deal with the loading and scrolling speeds.

Image Analysis Threshold won't work in program program freezes

I am a new user to labview and I probably have an error in my code but the only problem is that there are no error messages I get from my code and it runs up to a point at which the program freezes and I can't click any buttons on the Front Panel. Basically everything works up to when I want to choose the Threshold of my image at this stage it creates an image of the values in the threshold buttons but I can't adjust the values anymore to find out the best values.
I should say that I have tried my Threshold code on its own and it seems to work fine so I am really confused as to what is getting muddled up!
Any help on what might be wrong would be helpful. I have added the VI and an image that I am analyzing.
The process currently goes as follows, you load the image, then choose a colour plane (usually green) that gets extracted. Then you draw a line on the border of the object and it rotates the image to make it straight and then you choose a ROI and it cuts the main object out. Finally at the point when you should be choosing a Threshold it hangs.
Thanks for any help!
Solved!
Go to Solution.
Attachments:
Image adjustment version 4a.vi ‏143 KB
Sample 2.jpg ‏2359 KB

Hi,
The root of your problem is in how you have used multiple event structures. Typically, a program should just have one event structure, which handles all events during the program. The vision functions were all working correctly!
From looking at what you are trying to do, I think it would be worth using a well known architecture called the state machine with events. I'll attach some code framework to show you what it might look like for you, you can adapt it to suit your program. I hope its enough for you to get an idea of what I'm trying to show you, it doesn't have any code in it yet, but I think it gets the idea across, and if you combine it with the link below on state machines, it may help you to set out your code in a better architecture.
A few links you may find helpful:
Event programming: http://www.ni.com/white-paper/3331/en
Caveats for using events: http://zone.ni.com/reference/en-XX/help/371361J-01/lvhowto/caveatsrecmndtnsevnts/
State machine: http://www.ni.com/white-paper/2926/en
let me know if you have any more questions.
Ian S
Applications Engineer CLD
National Instruments UK&Ireland
Attachments:
Event state machine example.vi ‏14 KB

Cannot edit KEN BURNS EFFECT or DURATION/SPEED of image????

When I select a still image, I'm unable to edit the Ken Burns Effect or anything else in the "Photo" tab. Why is this? It's driving me nuts. It works with one section of images but not the other. I've been able to edit before. Thanks!

It works with one section of images but not the other.
I am not sure what you mean by "section". Are you referring to some of your photos? While in the "Photos" view, what happens when you click on the photo you wish to edit? Do you see a button that says "Show Photo Settings"? If you do, click on that. That will bring up the panel for the Ken Burns effect as well as a few other editing features.

Can I speed up final cut and motion rendering using qmaster?

I have set up a cluster to speed up exporting via compressor, but I am wondering if I can use the same technique to speed up the rendering (not the exporting) in final cut pro, or motion (or other programs included in the suite for that matter)
I read that you can use an xgrid cluster to do this, but I don't have a mac server to use xgrid and would prefer not learning a new program if I can stick with qmaster.
Thanks

Since your camera is acquiring at the rate of 1.2 frames/sec, that means a frame is ready every 833 ms. This is a function of your camera and may or may not be variable. This is the fastest possible loop rate you can achieve. With that in mind, if your image analysis takes 400ms and the motor 200ms, you still have about 233ms of time left to wait for the next image to be ready.
Yes, by reducing the acquisition window size, you reduce the amount of data moved across the PCI bus to system memory, but the camera control how quickly frames are being sent so this will not speed up your acquistion. It will, however, decrease your processing time since you are processing fewer pixels.
Also, a ring acquisition would be the best acquisition mode to implement since you w
on't need to setup a new IMAQ session each time. If you are worried about not finishing a move before the next image is acquired, you can just use the Wait for Move Complete function in ValueMotion and then continue with the image acquisition.
Jack Arnold
Application Engineer
National Instruments

Speeding up image analysis,

Similar Messages

Maybe you are looking for