Intensity Graph control using VC++ 6.0?

LabView supports an "Intensity Graph" control. Is there a way to accomplish the same thing with CNiGraph, CWGraph or CNiGraph3D?

Thank you for you answer. It was very helpful. One thing missing is the "intensity slider" that is part of the LabView control. It allows the user to adjust the image display. The other problem I have is that the 3D graph axes labelling always looks big and clunky no matter what I use for font type/size. What I am trying to emulate is shown in the attachment. Basically I am translating some LabView code into C++.
Attachments:
Image1.bmp ‏203 KB

Similar Messages

Intensive slider control use may lead to LabVIEW hang and system instability when coupled to event loop "Value Change"

Occasionally, over the past year or two, I have noticed on different machines and versions of LabVIEW (8.0 and 8.5) that intensive operation of the LabVIEW slider control (holding down the left mouse button and shaking the mouse quickly back and forth to move the slider for several seconds) may cause LabVIEW to hang with high CPU usage (50% on my current test machine, which has a Pentium IV "hyperthreaded" processor, so 50% is the max that a single thread can take up). In some instances, after the LabVIEW process is forced to quit, the system itself becomes unstable (more high CPU usage) or hangs, up to and including the need for a hard restart. I have seen this behavior both when handling slider change events the "old fashioned" way with a while loop and a case structure; and also when using a while loop with an event structure handling the "Value Change" event for the slider.
Attached is an .rtf file dump of the front panel and block diagram of a simple test VI that replicates this problem in my hands (using an event structure). The loop count runs to at least several hundred iterations before the crash occurs.
At least for the "old fashioned" loop handling method, adding a forced wait of a few dozen msec per loop didn't seem to be a workaround. I haven't tried this idea with the event structure, but don't expect joy to follow. The possibility of a system hang discourages aggressive experimentation.
Is this a known issue? Can anyone else replicate it (carefully)? In any case, I'd welcome other suggestions for workarounds.
Thanks,
Darin Nelson
Attachments:
Hang Slider.rtf ‏683 KB

Thanks for all your responses, and especially to those who are waggling their mice around trying to reproduce this. I'll respond to (most aspects of) the last several posts in one message.
LabViewGuruWannabe:
I'll see if I can get the original code to post in a separate message, I posted the .rtf originally because I thought there might be filtering on forum posts (I'm sort of new here). But I doubt you re-created it wrong.
The "more modern" test hardware was an Intel Core2 , 2.4 GHz, 2Gb RAM, on an Intel 965 family motherboard. As I said, it took me two tries to trigger the error on this hardware; in the case where it triggered, I was at almost 12000 iterations before things went south. 10000 iterations, therefore, might not be enough--or you might never get it, for all I know. My Pentium 4 Hyperthreaded machine (3.2 GHz, 2 Gb RAM) triggers the hang much more easily (always within 2500 iterations, usually within 1000), so there seems to be some hardware dependence here. Maybe the workaround is just to use hardware recent enough that no one not under the influence of strange compulsions will accidentally trigger the hang.
TonP:
After reading your suggestion, I tried something even more radical -- getting rid of the handling code altogether. I took the slider node out of the while loop, and removed all code inside the "Slide" case of the event structure.
On my Pentium 4 machine, the hang triggered after 677 iterations.
Brian Beal:

Huh, I never even noticed that checkbox existed. Anyway, yes, I
kept the default to lock out the front panel while the event process. If I
disable it… I still get the hang.
Message Edited by dman on 02-04-2008 06:36 AM

Cannot make vertical axis of intensity graph logarithmic

Has anyone else noticed that that the vertical axis on intensity charts cannot be made logarithmic? Labview gives you the option, but doesn't do anything when you try and check the box. Property nodes don't seem to work either. Tried this in LabVIEW 2011 SP1 and 2013. Am I doing something wrong or is this just a feature that isn't avaiable? I am just wondering if its my install or something that is really an issue.
My application is to plot the power spectrum of sound on a intensity chart to show history of sound recorded by a microphone in real time. Really good to pin point source frequencies of feedback.
Amit

Hi Amit,
This is a documented problem with LabVIEW and it is currently being worked on. Here are a couple other Forums that discuss it and offer workarounds for this issue.
Log scale in an Intensity graph
Possible bug in Intensity Graph while using interpolate colors on a Log scale
Intensity Graph - Mapping Mode
As shown in the 'Possible Bug' link above, a Corrective Action Request has been filed under the number 453407 as soon as a couple of weeks ago. To fully answer your question, it is not your install or anything like that, just an issue NI knows about and will be working on for future releases.
Douglas C.
Applications Engineer
National Instruments

BCG Brightness Contrast Gamma - Looking for BCG function returning color table / lookup table coded in pure G and applicable to Intensity Graph or Picture Control

Looking for BCG function returning color table / lookup table coded in
pure G and applicable to 8-bit Intensity Graph or Picture Control. Do
not want to use IMAQ Vision is this particular application. Any
help is appreciated.
Sincerely,
Don

Take a look at this color model software I put together. (Don't concern yourself much with the basic programming at this point.). The color component generator subVI is borrowed from a LabVIEW example.
First look at test_color2.vi. If you adjust the sliders of the various controls, you will see how it works. You can see that for pure B&W images, brightness, contrast, and gamma (which I think I have coded correctly) are all fairly straightforward to modify.
Now look at test_color3.vi. Here we try to adjust BCG for color images using the color model shown here where current color component clusters are carried into the BCG event cases via shift registers. You can see that once color is introduced into the image, it becomes difficult to mix the colorizing with the BCG model. And this is pretty much where I am right now.
Would be interested in comments / suggestions.
Sincerely,
Don
Attachments:
test_color2.llb ‏2251 KB

Control grid intensity graph

Hi,
I want to control the grid of an intensity graph to be exactly spaced by "1"... How to do?
See attached example in LV8.5. Start with default values and you get grid lines with spacing of 1.
Now change "size" to 15 and run again. Gridlines are now spaced by 0.5. Size=10 -> spacing of 0.2, size=4 -> spacing=0.1 ...
I want to have a spacing of 1 regardless of size parameter! How to achieve this?
Btw. I need both major and minor grid, as array size can grow to more than can be handled by major grid alone...
Message Edited by GerdW on 02-26-2008 03:09 PM
Best regards,
GerdW
CLAD, using 2009SP1 + LV2011SP1 + LV2014SP1 on WinXP+Win7+cRIO
Kudos are welcome
Attachments:
Test_Grid_LV85.vi ‏16 KB

Hi Simon,
thanks for clarification.
To summarize: major increment has to be bigger than minor and you have to set them both...
Scale labelling is not so important for me, but it would be nice to have each number when space is sufficient (in the example vi for sizes upto 15). It seems I have to change my graph formatting a little bit: make minor grid transparent when not needed (for small sizes) and switch gridlines on, when scale space is not sufficient to show all numbers...
In the attached example pic (today's measurement of pressure sensor wafer ) I use the gridlines to separate single chips. And I always want to separate single chips regardless of array size (=wafer/chip) - which can vary from 10 to ~80...
Message Edited by GerdW on 02-26-2008 05:26 PM
Best regards,
GerdW
CLAD, using 2009SP1 + LV2011SP1 + LV2014SP1 on WinXP+Win7+cRIO
Kudos are welcome
Attachments:
grid-example.png ‏27 KB

Using Excel with Visual C++ 6.0 (need a graph control that behaves as much like the one in CVI as possible)

I hate to beat this to death but I was unable to find a
clear answer to this question. Does NI provide Excel control from within
Visual C++ 6.0?
I read that NI supports the ANSI-C library only for Visual C++ 6.0 users:
Thread : "Re: Benifits of using measurment
studio for VC++ 6.0?"
NI REP : drohacek
Quote : "we made the decision to support Visual
C++ 6.0 users only through the ANSI-C interface and not through the Measurement
Studio MFC-based class libraries."
If there is a suggested way of controlling Excel from within Measurement Studio
for Visual C++ 6.0 then I’d love to know it.
If not, can a plain statement be made basically stating that if you want
to control Excel you can’t use a Visual C++ 6.0 environment even with Measurement
Studio support?
With .NET of course you can just decide to have Excel support added during
project creation. CVI can act as an ActiveX server and easily control
Excel just by building off the examples shipped with CVI. I see that
there is support, using the Measurement Studio for Visual C++ 6.0 Project Wizard,
for taking existing CVI projects and converting them to Visual C++ 6.0 projects
or for calling CVI libraries from a .dll from within Visual C++ 6.0. I suppose I could do the work
in CVI and then convert the project but I'm so deeply tied into all my MFC calls that I don't see
how I can cleanly include the CVI libraries into my existing VC++ 6.0 projects.
What I'm really after here is a visual graph control like the one in CVI that I
can use from Visual C++ 6.0. I purchased a 3rd party graph control for use
in VC++ 6.0 that works well but isn't really visual. I mean you can't
enter any values in it until you run the program and fill it out programmatically.
Then you can see which columns are two narrow, quit the program, adjust the
column width of your now empty control, and repeat, until you get the thing
looking the way you want. If I could call into Excel from within Visual
C++ 6.0 the way I do from within .NET
then I could use Excel to hold the table and just read in the values into my table
control at run time. Basically I'd use Excel as a visual development tool
for all my tables.
I do all my coding from within CVI and Measurement Studio for Visual C++
6.0. I'd use CVI for everything if I didn't depend so heavily on certain
outside controlled C++ .dll's. Could you please suggest then what I can
do to get Excel support for Visual C++ 6.0?
Last question :
Is there any plan to ever have a NI table control like the CVI table control for
use in Visual Studio? One that you can
completely set up before you do any compiling?
I imagine that the way Microsoft sets up their environment makes this idea impossible. Otherwise someone would
have come up with a truly “visual” table control before now.
Thanks,
Grant
Grant M. Johnson
Project Engineer
LECO Corporation

Hello Grant
You can most certainly use Excel with VC++ 6.0. Measurement Studio won't provided you with any classes to talk to Excel in VC++ 6.0. This is because Excel allows control via Automation and you can use its automation API to do anything you need with Excel. This is what CVI uses as well.
You can see this MSDN article on how to set this up. This mentions VC.NET, but it should work the same way.
Here is one that talks about VC 6.0 and Excel.
Here is a code project article about this.
Even with .NET, you have to do thru the Excel Automation support. Excel started shipping with Primary Interop Assemblies (PIA) which are .NET wrappers about the Excel Automation object model. C++ Automation is definetely not as nice as .NET, BTW
See this document for more information about the Excel object model.
I would not recommend using the CVI Excel libraries if all you want to use them for is Excel automation. You will end up creating un-necessary dependencies and go through extra layers that way. You can make calls straight to Excel from VC++ without requiring CVI.
One quick observation about your excel approach. If you decide to use Excel as your table, you might be requiring everyone who uses your application to have Excel installed on their machines. Just wanted to make sure this was acceptable to you.
Have you looked at the Datagrid Activex control, which is one of the common controls that ships with Visual Studio? You can add it to Visual Studio by right-clicking and picking it from the list of installed activex control. If you have not already, you should check it out.
Microsoft has made significant improvements in the number of controls they provide with .NET. They have a Datagrid control that seems to be what you need.
Measurement Studio did add some high level classes for Excel and Word Automation that simplify some common tasks, but these exists for VS 2003 C++ and VS 2005 C++, not for VC 6.0. Underneath, we end up using the same Excel automation classes, so you can easily setup something similar for VC 6.0. Plus you can find alot of references online on how to use the Excel Automation object model with C++.
We currently have no plans in Measurement Studio to create a table control unfortunately. I am assuming when you said graph, you actually meant table, since Measurement Studio already provides a ActiveX graph controls for VC++ 6.0 that is very similar to the CVI graph.
On a side note, VC++ 6.0 is really really old. Have you considered upgrading?
To summarize
- Yes, you can use Excel with Vc++ 6.0 without mstudio.
- Try using the Datagrid Activex control if you just need a table.
- Measurement Studio provides high-level excel and word classes for VC++ 6.0
- Measurement Studio provides a graph control for VC++ 6.0 which is very similar to the CVI graph.
- Measuremnt Studio does not have a table control.
- VC++ 6.0 is really really old. Have you considered upgrading?
Bilal Durrani
NI

Is it possible to delete a plot with using mouse to select it on the graph control?

For example, when I use PlotLine to plot thousands of lines on the graph control, I want to delete some of them, but there is thousands of plotHandle generated.
How can I get the plotHandle of the specific line on the graph control with using mouse to select it?
If I record information of the lines, ex. the plotHandle and the coordinates, on a table or list, etc. , I have to compare the information of lines to get the plotHandle of the specific line.
Is there any user-friendly way to delete a plot by select it on the graph control?
Solved!
Go to Solution.

Sure! Look at the example located in \samples\userint\graphcursors for an example of how to accomplish this task.
Briefly:
The cursor must be in "Snap to point" mode
(To be confirmed) The graph must be in "Retain" data mode
You must have a callback for the graph control
You can use GetGraphCursorIndex to retrieve the plot handle
GetGrpahCursor retrieves values of cursor position in engineering units
Proud to use LW/CVI from 3.1 on.
My contributions to the Developer Zone Community
If I have helped you, why not giving me a kudos?

LabView crash when intensity graph is in P6 of tab control

Any available fixes for LabView crash when intensity graph is put on page 6 of tab control and right-clicked??

Michael,
This has been fixed in LabVIEW 6.1.
Regards,
Cyril Bouton
Applications Engineer
National Instruments
Cyril Bouton
Active LabVIEW Developper

Using intensity graph

Hi,
Currently, I have some problem using the intensity graphs. I am using the Time frequency toolbox in Labview. Currently, I am having a problem where I needed to plot the spectrogram of the signal. I have tried to maintain a plot of spectrogram at 0 to 4s and another at 4s to 8s. I am trying hard to plot two graph together, to form a plot of 0s to 8s, in the spectrogram. Furthermore, the time axis will continue to increases at 4s interval with updated spectrogram. I wonder how could i achieved in having two spectrogram with two different time interval to display a common plot? Thanks a lot for advice.
Rdgs
kengkc
Solved!
Go to Solution.

Hi kengkc,
Correct me if I am wrong, but are you saying that you have the data for the first 4 seconds of your acquisition (t = 0 to 4) and the next 4 seconds of your acquisition (t = 4 to 8) and you want to combine them into one plot? I presume since you were able to plot them in an intensity graph to begin with, that they are already in arrays.
If that is what you want to do, you can easily concatenate your arrays using the Build Array function (this means to append one array to the end of another). Add your Build Array function from the Array subpalette, and make sure you right click it and choose Concatenate Arrays so that you dont get more dimensions in your resultant array. Then expand the Build Array function so that you can wire your arrays into it.
Please refer to the attached screenshot for an idea of what I mean.
Hope this helps.
Best regards,
Vern Yew
Applications Engineer
Attachments:
concatenate.JPG ‏17 KB

Image processing from .txt file onto an intensity graph

I am doing a mini project in my class and I was wondering if anyone could help me. It about image processing but I am bit stuck.
Heres the idea:
"An image is really nothing more than a 2D array of data. The value of every element in the array corresponds to the brightness of the image at that point.
In this project you will create a VI which loads a 2D array of data and then displays it on the screen using the Intensity Graph. Three example files (boats.txt, gordon.txt and parrot.txt) are available on the module webpage that you can use. However, you can also use any other black and white image you like, but will need to convert it to a “text image” first. To do this you can use some software called “ImageJ” which is available on the computers and is free to download.
You can vary the brightness of an image by adding the same value to every element in the array.
The contrast of an image is adjusted by multiplying every element in the array by the same value. Using numerical controls and simple array mathematics, you should adjust the brightness and contrast of your displayed image.
Some other ideas that you could try with image manipulation are:
o Invert an image (change black to white and white to black)"
First I am having problems putting my picture onto the graph. It is in the write file but the colours are not correct and the image has rotated 90 degrees. I will upload my VI so far when I get back onto my computer.
Would really appricate the help! Thank you for reading

Hi charlthedancer,
Here is an example to get you started.
Kind regards,
GajanS
Attachments:
Test.vi ‏14 KB

Programmatically disabling/enabling cursor in intensity graph

Hello all,
I am building a LV interface (LV2011) for a stepper XY stage. The system "scans" a surface, displaying it on the graph. I set up to cursors to zoom into a region of interest (ROI). Pressing a button resizes the scales to the ROI (zooming in on the area). Sections of code move the XY stage, and read a device. When done I want to zoom into a ROI and use a THIRD cursor to manually move the XY stage. I have this working with TWO graphs, based on the "region of interest.vi" I found in the dev zone.
What I would like to do is have a boolean toggle controlling ONE GRAPH between setting min/max (cursors 0 and 1) and manual stepper position (cursor 2). If cursor 2 is visible I want cursors 0 & 1 not showing and visa versa. Can I do this?
I attached the example - bascially what I did was add a cursor to the 2nd graph, grabbing the XY vales and sending to the stepper controllers. Works great! I am trying to "combine the two" by resizing the orig graph with a button click (works fine) and then click another button to show the XY stage cursor (and hiding the min/max cursors) - thats the brick wall
On a seond note - controlling a XY stage using the intensity graph's cursor is really nice; however it would be great if I can control the cursor similar to a slider's "coerce" - has anyone done this? I assume I have to write a routine in the event that performs some modulo function. Just curious!
Thanks,
Steve
Solved!
Go to Solution.
Attachments:
roi.llb ‏40 KB

Your exmaple works perfect. Howver I am having probs with Val(sngl) on the XY graph. On the example you gave, lets say I move the cursor with the mouse - event fires, everyone is happy. If I move the cursor with CODE - nada. I even read the "Value" of the graph and wrote it to "Val(sngl)" to try to fire the event. How does one force a trigger on an XY chart when you move a cursor (say cursor idx2) using code?
Better yet - why when I read the Val and write to Val(sngl) does the event not fire??? I have been searching for a basic XY table with sensor -type project but havent found one yet. Everyting works great - draginng the cursor moves the XY stage. Now I need to do the auto-scan part Use code to place the cursor - trigger the even that moves the stage.
Thanks
Steve
Attachments:
XY_Stage_Rev8.vi ‏53 KB

Change colors of individual pixels on intensity graph?

I would like to be able to change the color of a set of individual pixels within an intensity graph, to blue for example, while leaving the color table otherwise as is, for example, gray scale. The change to blue would be based on location, not on current gray value. I think I can accomplish this using the mask fx with picture controls, and then convert to intensity graph. Is there a more direct way? A good example to demonstrate this capability would be just using a 'mouse down' event on an intensity graph to change the color at the mouse location. Any thoughts are appreciated.
Sincerely,
Don

Yes, seems to work fine .
If you have LabVIEW 8.0, there is another possibility. There is a new property that allows drawing any pictures or picture elements on your graph area. Of course you could draw individual pixels, but nothing prevents you from e.g. drawing circles, squares, odd shaped areas, arrows or event text. ... Anything you could do with a plain picture indicator. One example that ships with LabVIEW 8.0 is Envelope.vi, see image.
Message Edited by altenbach on 06-12-2006 07:31 PM
LabVIEW Champion . Do more with less code and in less time .
Attachments:
envelope.png ‏32 KB

SubVI - Main VI: Intensity graph update too slow ??

I have an intensity graph in my subvi that I call onto my front panel
using control references. It works nice but it is a lot slower than if
I just call it as a subVI which loads upon call. Has anyone faced this
problem before ?
Thanks
Kudos always welcome for helpful posts

I agree with tst, your PP VI runs at unlimited speed for no good reason. You should pace it with a small delay so it will release the control to other processes after each iteration.
Also your DAQ part seems inefficient. Why do you configure the DAQ task from scratch, acquire, and close at every iteration? This is way too much work. You should flatten this out to the PP vi. Configure outside the loop on the left and close outside the loop on the right. Place only the ACQ part inside the loop.
Also, place the real terminal of the graph inside the loop and clear it once at the beginning using a local variable. A local variable is always more expensive and less efficient.
I don't understand why you need a timeout event in the main VI, it does not do anything except spin the loop, right? Just delete the timeout event.
LabVIEW Champion . Do more with less code and in less time .

Region of Interest (ROI) on Intensity Graph

Hello,
I hope somebody can help me. I have read the other posts available regarding choosing ROIs on intensity graphs but they don't really answer my question.
I am taking a series of still images. Having stored the image as an array and displayed it on the intensity graph, I need some way of selecting a rectangular ROI (preferably with mouse) to define an array subset and performing max/min/mean/standard deviation measurements on this region. I am aware that the array would need reshaping to work with the in built VI for mean and standard deviation but cannot seem to find a way to may the selection of the array subset in the first place.
I would be very grateful for any pointers
Sarah Bohndiek

Hi,
since you're using an intensity graph, you could use an event structure to capture for the mouse down and mouse up events.
They both give you the coords of where the mouse was. The nex trick would be to convert that into array position co-ordinates, which requires knowing how the data is presented in the graph (actual graph area size, and the amount of data passed to it.) It's easy enough to get the size of the plot area, and you can get the position of the control itself, however calculating exactly where the plot area is inside of the control is not available.
If you plot area size is fixed, you could use a picture control, set to transparent over the top of the plot area. Then use the same trick for the event returning co-ordinates. If you change the values on the scale however, the actual plot area can change in size too (since the renumbering on an auto-scale can cause the scale on the axis to move if the text increases or decreases in length)
By far the easiest route would be to buy the vision development module and use the image control which has the built in ROI capabilities (and editing of the ROI).
Hope that helps
Sacha Emery
National Instruments (UK)
// it takes almost no time to rate an answer
Attachments:
roi on intensity (picture).vi ‏73 KB

Open 32-bit jpeg & convert to intensity graph form (LV8.5 & Vision 8.5)

I will cross-post this in the Machine Vision forum as
well. I have an LLB to open images with IMAQ vision utilities, and
then attempt to convert to intensity graph format. I obtain the color
table of the image from a Vision utility, and subsequently apply to the
intensity graph in hopes of duplicating the vision container contents.
I am unsuccessful in converting 32-bit jpeg and was hoping for some
help. Please let me know if this in fact is even possible to do.
Attached is an example jpeg image file.
Sincerely,
Don
Attachments:
open_image_file_and_convert_to_intensity_graph1.llb ‏207 KB
Photo_111107_0021.jpg ‏240 KB

DonRoth wrote:
Hi George:
Yes - I understand I am getting only one color plane.
With LabVIEW / Vision only, can this jpeg image be displayed in an intensity graph with the proper (original) color information? You say the LabVIEW Read Jpeg File.vi reads this image fine - but how do you display the output of this fx in an intensity graph with the original / proper color information (again using only LabVIEW / Vision, or maybe your toolkit)?
Unless it is an 8 bit color image already you can always down sample it to 8 bit. I think the IMAQ Cast function should do that, otherwise there is for sure some function to change color depth. Colors will of course change slightly as it is very unlikely that a 24bit image especially if it is a photo, has only 256 unique colors used.
Once you converted your IMAQ image into an 8 bit image extracting the bitmap and color dtable is easy and then applying it to an intensity graph of course too. Remains the question why you wouldn't want to use the IMAQ control. If it is about IMAQ license costs then of course you won't be able to use other IMAQ functions either so this advice won't help then.
Rolf Kalbemratter
Rolf Kalbermatter
CIT Engineering Netherlands
a division of Test & Measurement Solutions

Intensity Graph control using VC++ 6.0?

Similar Messages

Maybe you are looking for