Linescan camera

Hi
I am using a Atmel Aviiva SM2 Linescan camera with a PCI-1426 frame grabber - I am operating it in triggering mode with an external trigger of 10Khz and exposure times of
20 us,40 us & 80us - The height of acquisition is set to 1 - The number of frames acquired however doesn't change at all - it is around 12000 frames/sec (shown on MAX) - I am not sure how the no. frames grabbed is calculated - can anybody help me out .
Suddha

Suddha,
I am not sure how the number of frames grabbed is calculated, that is something low level in our software. If you change the rate of your trigger source, does the number of frames acquired change accordingly?
Christopher W.

Similar Messages

XML file for GigE Vision camera

Hello,
I am working on a design of GigE Vision camera. It should be very simple linescan camera. I implemented the whole required GigE Vision register set, can communicate with the camera in MAX. The problem is that I need to creata this XML file for the camera.There is this XML file here:
http://www.emva.org/cms/upload/Standards/GenICam_Downloads/SFNC_Reference_Version_2_0_0_Schema_1_1.x...
What should I do with it? My first version of the camera will have 2 funcions: turn it on and turn it off I am bit confused, because the xml file template has lots of functions, that I don't really need. Should I delete them? Or leave inside with default values? What does minimal configuration XML file need?
Regards,
Linus
Solved!
Go to Solution.

linru wrote:
Thank you very much! Perhaps I was too concentrated on digging through various documents and missed the important info.
And one more question: how does NI software build the *.icd file? By reading XML file? Or by reading registers in my camera?
Both. The XML is processed by the GenApi software component which then translates it to register operations. The XML file thus controls what features are visible based both on the XML file as well as the camera registers (features in the XML may be conditionally available based on the <pIsImplemented< tag). Next, the IMAQdx ICD file contains the subset of features in the XML file that are available and are tagged as <Streamable>, meaning they can be saved to a settings file.
Eric

Question about camera file generator

Dear all,
I was trying to set up my camera with the frame grabber using camera file generator. My camera is a linescan camera, which is 1024x1 pixels.
My question is, in the camera file generator, I can choose to display maximum number of rows which could be any number I key in, for example, I can choose to display 1024x256 or 1024x800.So since my camera has only one line of pixels, where are all the other rows come from?How does it work?
Thanks a lot!

Hi
When you create your icd file, the height value, will be the maximum number of lines you can accumulate.
So when you will use your camera with MAX or LabVIEW, you can specify an image height within 1 and this value.
I suggest to set this value pretty high (10000 lines for instance) so that you are free to request a big image in your application.
Regards

How can i use PCIe 1433 and basler line camera to construct image and pocess it?

hello! every one ,I am new to machine vision.i have two problems to ask.First,how can i use line camera and 1433 to acquire 2-D image?second,if i want to pocess the 1-D signal ,how can i do it ?i expect for your answer ,thank you!

What you are trying to do is not something you can easily do in a forum, but I can give you the basics.
You have a CameraLink frame Grabber (PCIe 1433) and a LineScan Camera
The framegrabber needs to be configured to understand how to communicate with the camera. You need a file from the camera vendor. This file might already be installed. You can use MAX to check if it is, under Devices/IMAQ.
Assuming you can get to the point where you can communicate with the camera, you need to configure your acquisition.
Since the camera is a line scan camera, it captures image data one pixel ine at a time. You will need to configure the frame grabber to determine how many lines per image you want. The low limit is usually 1, and the high limit is either based on the framegrabber's internal memory, or that of the computer.
If you are using labview, you will need to open a sessing to the camera, acquire the image(s), and then close the session when you are done.
Once the image(s) are captured, you can performa analysis using the NI Vision Toolkit.
Sorry, but there is no way to condense years into one post.
Machine Vision, Robotics, Embedded Systems, Surveillance
www.movimed.com - Custom Imaging Solutions

Non linearity between pixels and widht for linescan

Hi
I'm developing a width gauge based on a linescan camera.
The gauge is measuring the surface of a steel strip moving at 200m/sec.
I found a non linearity between the relation of pixels and real widht:
1904mm X 6635,6pixels
1308mm X 4616,7pixels
814mm X 2880,6pixels
I've tried the working distances between 2600 to 2800mm.
The field of view is 2200mm.
The focal lengh is 35mm.
The size of CCD is 28.67mm (3.5um X 8192pixels - Basler raL8192-12gm).
Somebody already faced this problem?
Thanks,
Alexandre.
Solved!
Go to Solution.

Hi Greg
Thanks for your idea. I implemented a polynomial from Excel according the attached. It's running well until now.
Another problem that I'm facing is the oscilation on the image (https://www.dropbox.com/s/cboyglx74g8sd93/width%20gauge.mp4?dl=0) when the grid and filter are applied. I would like get stable the filtered image and grid from edge detector.
Does anyone know how to solve this?
Thanks for the moment.
Attachments:
calibrcurve_ml2.xlsx ‏20 KB

Calculating motion rate for given line scan rate

Dear friends,
I need to understand the calculation of motion rate for given line scan rate.
i have line-scan camera connected to servo motor.
i know that the line-scan camera can run at 7.2 khz line rate (7200 lines in 1 second).
Each pixel need to be 0.1mm (in the direction of motion). 10 lines of scanning is 1mm.
i need to scan image of object that its length is 400mm (0.1mm*4000lines)
First, the Conveyor stop, the object that i need to test with the linescan camera is now 5cm from the camera.
i want that the servo motor will move the conveyor for 45cm and stop.
(the conveyor accellerate, move const speed, decellerate, stop. the line scan camera connected to 10khz encoder for each revolution)
Now i'm getting lost with the calculation.....
if 1pixel = 0.1mm , then i need the conveyor belt be able to move 0.1mm for each pulse of the encoder???
i want that the servo motor will rotate chain (like the attached image),
but i dont know how i need to calculate the gear diameter, the chain pitch, etc...
and specially, how this parameters influence the line scan rate, and what is the role of the encoder here...
Thanks for any support.
Moti
Attachments:
Chain_gear.jpg ‏17 KB

It sounds like you are mixing two methods of using a line scan camera.
The first method is to run the camera at a fixed speed. If the speed of your belt is constant, this can work okay. To calculate the desired speed, let B=belt speed (mm/sec), L=line rate (lines/sec), R=mm/line (0.1mm/line in your case)
B = L / R This equation can be rearranged to calculate any of the quantities.
The second method is to use an encoder to trigger each line captured by the camera. In this case, the speed of the belt is not important as long as you don't go too fast for the camera. In this case, you need to know how many mm the belt will travel for each pulse. To calculate this, you would probably measure how far the belt travels during one revolution of the encoder. You should also know the number of encoder pulses per revolution. Dividing distance by pulses gives you distance per pulse. Since you want 0.1 mm/pulse, you would need to adjust either the number of pulses in the encoder or the diameter of the belt roller. If E=encoder pulses and D=belt distance, R = D / E.
Bruce
Bruce Ammons
Ammons Engineering

Calculating motion rate and conveyor move in mm for every pulse of encoder

Dear friends,
I need to understand the calculation of motion rate for given line scan rate.
i have line-scan camera connected to servo motor.
i know that the line-scan camera can run at 7.2 khz line rate (7200 lines in 1 second).
Each pixel need to be 0.1mm (in the direction of motion). 10 lines of scanning is 1mm.
i need to scan image of object that its length is 400mm (0.1mm*4000lines)
First, the Conveyor stop, the object that i need to test with the linescan camera is now 5cm from the camera.
i want that the servo motor will move the conveyor for 45cm and stop.
(the conveyor accellerate, move const speed, decellerate, stop. the line scan camera connected to 10khz encoder for each revolution)
Now i'm getting lost with the calculation.....
if 1pixel = 0.1mm , then i need the conveyor belt be able to move 0.1mm for each pulse of the encoder???
i want that the servo motor will rotate chain (like the attached image),
but i dont know how i need to calculate the gear diameter, the chain pitch, etc...
and specially, how this parameters influence the line scan rate, and what is the role of the encoder here...
Thanks for any support.
Moti
Attachments:
Chain_gear.jpg ‏17 KB

Discussion continued here:
http://forums.ni.com/ni/board/message?board.id=200&thread.id=20051
~~~~~~~~~~~~~~~~~~~~~~~~~~
"It’s the questions that drive us.”
~~~~~~~~~~~~~~~~~~~~~~~~~~

Can't include library .vi in vision deployment

Hi,
I'm deploying a vision application (LabVIEW 7.1.1, Vision development module 7.1.1)
I have both the LabVIEW and Vision run time engines at 7.1.1
When running the installed application on the target PC many NI vi's used are not found.
Most are solved by including libraries in the build, but one isn't.
imgCopyAreaWithOccur.vi (used in IMAQGrabAcquire.vi) can be found in DLLcalls.llb.
However, if I include either imgCopyAreaWithOccur.vi or the full library DLLcalls.llb in the installer, the vi is still listed as missing.
Any ideas where I'm going wrong?
Thanks
Bandit.

Hi Tom, Thanks for your reply. It's turned up an anomaly and I don't know why
On my development machine IMAQ is at 3.5.0, but on the target machine it's at 3.1.0.
I don't actually want to go as far as 3.5.0 as it doesn't support Win NT. In the end one of the targets for my app is an NT4.0 machine.
The thing is I don't know why my development machine has IMAQ 3.5.0, as the vision development module 7.1.1 CD installed to it only has IMAQ 3.1.0.
I should probably try to roll back my development PC to using IMAQ 3.1.0 - can you advise if this is likley to cause me any difficulties?
As for system info, the situation is as follows.
We have a Dalsa Spyder linescan camera we have used for a while, with a PCI 1424 acquisition card on a windows NT system.
We then got a Dalsa Piranha camera and a PCI 1428 card, which is paired with a Win XP system.
My current task is to upgrade my linescan application to enable the use of either camera (done) and make it distributable (doing that now).
The Piranha becomes the main system (it's faster) and the Spyder gets mothballed. Should we have enough work to need 2 camera's we then purchase a second license and bring that system back into use.
Let me know if you need any further info. I have this thread set to notify my e-mail account inthe event of a reply.
Regards,
Bandit

Consumer producer - Acquisiton image but only producer loop run

Dear all
I try acquire and save image from Basler camera L301kc and NI PCe 1427 framegrabber. I create a producer/ consumer loop because the save take more time that the acquisition and if I do both in the same loop I lose frames. But I have problem: only producer loop run.
I have read all the threads but I didn't manage to solve my problem.
I have attached a VI code of how I was trying to do it. I am using LabView 2011.
Please any comments or help is deeply appreciated.
Thank in advance
XuanThuy
Solved!
Go to Solution.
Attachments:
Acquisition Image for Sample 03.vi ‏75 KB

Dear Kira T,
The producer and consumer loop has run but the Image Out in consumer loop is empty. Bruce attention to the error wire from the consumer loop wired to the destroy queue vi. But I don't find error.
I have attached a new VI code.
Can you help me ?
Thanks and regards
XuanThuy
Attachments:
Acquisition Image for Sample 03.vi ‏71 KB
Linescan camera setup Sub VI.vi ‏26 KB

8-bit image acts like a 16!

Hi all, Photoshop newbie here. We create images with a color linescan camera and save to a TIF file. They are all 8-bit images, no colorspace assigned.
I took 2 images within 6 minutes of each other. Photoshop (CS2) identifies both of them as 8-bit images. BUT, it treats one file like it's a 16-bit image with missing filters etc but is fine with the other image. The only difference I can see is the file size. The file that works is about 110MB, while the problem file is about 161MB.
Has anyone seen anything similar, or knows how Photoshop decides when an image is "filter-worthy" and when it's not? Thanks in advance!

What's interesting is that a simple resize of the image fixes the behavior of the filter menu. So going from say ~520 ppi to 400 ppi, all of a sudden you have access to the greyed-out filters. But you only have to do that on the larger of the 2 files. They both have the same native resolution (we scan at 2048 pixels/line at 200 lines/cm).
It's also cross-platform, same problem for Macs and PC's running the same version of Photoshop. Strange...

Need help with line-by-line triggering

We have a Basler L104k-2k digital linescan camera interfaced to a NI PCI-1428 frame grabber. We are writing our own custom software using VB6 and NI-IMAQ. We are trying to trigger the image acquisition off of our table motion so that slowing down the table will slow down the shutter speed, etc. We are controlling motion with a NI PCI-7342 servo card and have a RTSI cable connecting the 7342 and 1428 cards. We have been unable to acquire images by line-by-line triggering via the RTSI signal. We have looked at all of the examples we could find on the NI site and still have not been able to get this to work. We are setting up a signal in IMAQ to trigger off of the encoder and route it over RTSI, with some skiptrigger value (say 8000).
We only seem to be getting one line diplayed in the video image. We have used the basler camera setup tool to set the camera to ExSync, Level Control.
We are only allocating one image buffer in IMAQ. We are assuming that the grame grabber takes all of the individual linescans and composes one image out of them. In MAX, we have set the image height to 2048 lines. Any ideas? We have hit the wall on this.

Attached is an example VI for acquiring images from a line scan camera, triggering the start of each image and the individual lines. This is not the most elegant of examples. You will need to declare the appropriate number of buffers in MAX before using the VI. The image is built in PC memory. For example, if your image is 2048 x 10000, then you will need 10000 buffers. If you have any questions, please feel free to contact Graftek Imaging, cbr 512-416-1099.
-Robert Eastlund
Robert Eastlund
Graftek Imaging, Inc.
Phone: (512) 416-1099 x101
Email: [email protected]
Attachments:
Linescan_-_Trigger_Start_Acquisition_and_Line.vi ‏86 KB

1424 does not pass diagnostics after running program

Hello, we are using the PCI-1424 card with a Dalsa linescan camera, and are having trouble trying to do ring acquisitions with different buffer sizes. The camera either times out or always returns the number of lines that it is first set to, ie; if the original number of lines is 100, and the number of lines is then set to 1000 (and confirmed using imgGetAttribute(DisplaySid, IMG_ATTR_ACQWINDOW_HEIGHT, &nBufSize).
I've noticed that the 1424 board does not pass 2 of the diagnostic tests in MAX. The details it returns are:
Advanced Control Circuitry failed
Configuration could not be written to when Confwren bit was set. [F1.5]
Timing Circuitry failed
Counter configuration could not be loaded. [T1.0
Is this a programming error on my part? I am closing the session and the interface, which in my opinion should be all that is needed to reset the board to the default settings. This does not work, even the snap acquisition in MAX fails.

I'm not surprised at the behavior that you are seeing. When the acquisition is started the board is programmed with the acquisition size and when you change that it has to reprogram the board while it is acquiring, so either it will timeout or it just won't do it and you will only see the first size. I'm not sure why the diagnostics are failing but it sounds like you can reboot and that resolves the issue. I would recommend looking into a couple of alternatives. Variable Height Acquistion uses a hardware signal to indicate how many lines should be in an image. If you didn't have a signal you could generate one with a DAQ card or you could programmatically assert
and deassert one of the trigger lines on the 1424 but that would not be the most accurate depending on your line rate. The other alternative would be to set the image size to the largest size and then just crop the images.
Regards,
JR A.
Application Engineer
National Instruments

Save Bit Depth in MAX

I set Bit Depth to be 12 for my Linescan camera. It's OK when I view the image frame in MAX, but it is shown in 16 bit when I view it in Matlab. Is there any way that I can save it into 12 bit image data instead of 16 bit? The version of MAX is 4.5.
Thanks

You can not address anything smaller than a byte in your computer memory. So your 12 bit will be converted to 16 bit.
Besides which, my opinion is that Express VIs Carthage must be destroyed deleted
(Sorry no Labview "brag list" so far)

Error 1074397150 in gated IMAQ acquisition

Hi,
I have a SUI linescan camera which is read out with an NI 1427 framegrabber. I use an acquisition routine similar to LL ring acquisition. I want to gate the camera externally such that a fixed number of lines is acquired when the gate is high. So I put a TTL signal with duty cycle 50% on the gate of the camera. What should happen is that when I start grabbing the images from the camera, the buffer rate is half as high as without gating.
This is indeed the case SOMETIMES. In many cases labview doesn't start acquiring images and the acquisition 'while loop' is locked. The error is in IMAQ Extract Buffer->IMAQ Examine buffer->IMAQ img WaitForIMAQOccurence with error code 1074397150. Apparently the occurence is not made by the .dll imgSessionExamineBufferWithOccur This problem occurs even with a gating speed of >1 kHz, so it is no timeout issue I think. If I operate in a mode without gating, there is no problem and the acquisition rate is as expected.
Thanks in advance for the advise.
Sincerely,
Jeroen

Dear Jeroen,
Thank you for your post on our discussion forum. or some reason your video signal may be corrupt. If this is the case te following example may be suitable for you; http://sine.ni.com/devzone/cda/epd/p/id/3859 .
To test that your camera is returning valid images, try running some of the high level IMAQ examples that ship with LabVIEW by going to Help >> Find Examples. In the middle folder list, browse to Hardware Input and Output, and select the IMAQ folder. Browse to the High Level folder, and try running some of the examples in this folder to verify that the camera is working properly with LabVIEW.
I hope this helps.
Best regards,
Martijn
Martijn S
Applications Engineer
NI Netherlands

CCD interfacing problem

I need to create 2 square waves, the first (SI) having a freq. between
200-1000Hz (user selectable); the second sq wave (CLK) is a pulse train
(10-500kHz) of 64 pulses. The pulse train (CLK) is triggered by the first
square wave (SI). Its to interface a CCD linescan camera (64 pixels) to
LabVIEW using an ATMIO16E2 board. The first sq. wave specifies the charge
integration time (over which charge is accumulated), the pulse train
transfers the pixel charge to the analogue output. The output period is
triggered by the SI pulse.
Any ideas?
Leon

Tks for your answer. Actually it is interfacing other product or solutions, for example human resource solutions or from microsoft dynamics manufacturing module. is it possible to do that ? I appreciate your answer. TIA
Rgds,

Linescan camera

Similar Messages

Maybe you are looking for