How to do acquire different video lines

[VincentY]

Hi Daniel

I totally understand what you said.

But I still need to get all 525 line data from one ntsc frame during 1/30 second.

I tried to use the "any line" trigger.

The problem is I have to find which line is the first line.

And I'm not sure the sequence of each line is correct.

Could you please suggest a better way to complete it?

Thanks in advance.

Vincent

[dansch]

Hi Vincent,

I am not sure if I understand what you mean by having to find the first line and the sequence of lines potentially not being correct. Can you just trigger off the first line that comes into the digitizer and process the data in software to determine what you are looking for? Perhaps I am just not sure exactly what your data entails and what you are trying to acquire and/or process. Sorry for my confusion.

[VincentY]

Hi Daniel

Let me explain more detail.

I want to get all 525 scanline data in sequence from one ntsc frame during 1/30 second.

I use the "any line" video trigger to capture 525 scanlines then I get a array with 525 elements.

How can I find first scanline of a NTSC frame in the array?

And I am afraid of the first element of the array is corresponding to the 23th scanline.

But the second element of the array is corresponding to the 25th scanline. It should be the 24th scanline.

I hope you can understand.

Maybe I shouldn't use the "any line" video trigger.

Could you please suggest a better way to complete it?

Thanks!

Vincent

[Dennis_Knutson]

I don't understand why you a doing the 'random line' capture. Set it to 'Line Number' and pass 1 through 525 to the 'line number' input.

[VincentY]

Hi Dennis

If I use "Line number" to trigger, I have to change the "Line Nunber" in labview program.

So I have to wait at least 525(scanlines) * 1/30(frame per second) = 17.5 seconds to get all 525 scanlines data.

It's too long.

That's why I need to get all 525 scanline data in sequence from one ntsc frame during 1/30 second.

Do you have any idea??

Thanks!

Vincent

[Adam G]

Hello Vincent,

Have you tried triggering off the first line then processing your data in software?

Best Regards,

Adam G 

[VincentY]

Hello Adam

Yes, I tried any combination to get all 525 scanlines.

If "Line number" trigger is used, it will take a long time.

If "Any Line" trigger is used, I don't know which record is the first scanline.

I must process the data in software,

It takes a long time,too.

Is it possible to get all 525 scanline data in sequence from one ntsc frame during 1/30 second??

I think maybe it's hardware limitation of 5114.

Best Regards,

Vincent

[dansch]

Hi Vincent,

I think the reason that we are confused is that you keep mentioning that you will not have time to capture 525 scan lines in 1/30 seconds. You mentioned in a previous post that "525(scanlines) * 1/30(frame per second) = 17.5 seconds", but I am not sure what you mean by this. In reality, an NTSC frame consists of the 525 scan lines and the NTSC signal itself consists of 29.97 (~30) interlaced frames per second. Thus, the time that it takes for these 525 scan lines to occur is 1/30 of a second.This implies that a single scan line will exist for ~63.5 µs.

If you specify for the digitizer to trigger on line 1, it should trigger at the first line and begin taking samples. Instead of telling it to acquire the number of records for a single line, tell it to acquire what you would expect the entire frame to require. For example, if you are sampling at 100 MS/s, and you want to acquire an entire frame of 525 scan lines, you need to acquire for a total of 0.03333 seconds (1/30 seconds). This means that you need to take approximately 3,333,334 samples to get the entire frame. Is this similar to what you have attempted to do? Am I still misunderstanding what you need or what you are trying to get to happen? Sorry for all of the confusion on this, hopefully we will get it worked out asap.

Regards,

[VincentY]

Hi Daniel

I can specify for the digitizer to trigger on line 1 and take 3,333,334 samples at 100 MS/s.

But How can I I divide 3,333,334 samples into 525 scanlines?

Because the UUT is unstable, the first scanline maybe consist of 6100 samples and the second scanline maybe consist of 6500 samples.

That's why I need the hardware trigger to help me to seprate the scanlines from one frame.

"525(scanlines) * 1/30(frame per second) = 17.5 seconds" means if I use "Line number" trigger,I can get one scanline from one frame at most.

Because I have to change "Line number" for each scanline in software.

Thanks!

Vincent

[dansch · Accepted solution]

Ahhh okay I understand now what you mean, and I agree. Unfortunately, there is no good way to do this. The best you can possibly do is take all the 3M+ samples and separate them out in software. If you had a PXIe-5122, you could possibly stream the data to a FlexRIO module and process it in the FPGA, but there is no other good way to do that in hardware beyond this type of system, so it mut be done in software. If the scan lines were deterministic (this is the assumption I was originally going off of), it would not be too difficult to just divide the data evenly, but it sounds like that is not the case for you. Sorry for the bad news...

Regards,