Jump to content

TobyD

Members
  • Posts

    636
  • Joined

  • Last visited

  • Days Won

    3

Posts posted by TobyD

  1. QUOTE(menghuihantang @ Nov 1 2007, 09:22 AM)

    I was trying to process some fiber images. One of the steps is to get the skeletons of all fibers and count the numbers. But the skeletons are not very clean, small branches at the end. Usually imaging software has the trimming function to get rid of those, but I have difficulty finding such a function in LabVIEW vision, or LavVIEW doesn't have one at all? And another function which can return the numbers of neighbors for each pixel seems also unavailable in LavVIEW.

    Does anyone have any idea how to achieve these two functions? Appreciate it.

    I don't have a lot of experience with vision, but I know that once you convert the image to binary (it looks like you have already done this) there are multiple methods of trimming, filling, finding neighboring pixels, and counting individual objects. Unfortunately I don't have vision installed on my new laptop so I can't point you to anything specific.

  2. QUOTE(mross @ Nov 1 2007, 11:02 AM)

    Though as I was looking at it I kept thinking, I wouldn't trust that - running the loop every 1ms, clearly it works for lighting 3 booleans. But the loops aren't independent so if one light operation runs slow, it would hold up the others. Dataflow and all that.

    That would certainly be something to watch for. You could increase the delay, which would eliminate any of the problems you bring up, but each interation then has the potential to be off by as much the loop delay. It would probably be safer to go with a 5-10ms loop delay - especially for an application like this where the on/off times are measures in minutes. Incandescent bulbs usually have an illumination time of >15ms anyway.

  3. QUOTE(crelf @ Oct 30 2007, 03:21 PM)

    Not really a bug per se, but has the LAVA site slowed down dramatically in the last week or so? I thought it was the link at my hotel, but it seems to take ages, irrespetive of my link speed. The banner seems to load okay (takes 4 - 5 seconds), but then the rest takes anywhere between 15 and 45 seconds.

    I haven't noticed any significant lag. I'm browsing both from home (Comcast cable connection) and from work in WA state and the pages seem to load quickly.

  4. I've been preparing for my CLD exam and took the "Security System" sample exam yesterday. I fell like it went pretty well, but I thought I'd solicit feedback from the experts. After I completed my project I looked at the example solution and it was a significantly different approach. In fact, I feel like the example solution does not even meet the program requirements (i.e. an event is logged every time you push a button, not every time the status changes).

    Take a look if you have time and please don't hold back. I can handle criticism - I'm just trying to get it right. I'm looking for specific feedback in the areas that the exam will be graded in.

    Also, I know you cannot discuss the actual exam, but are there any specific areas I should focus on besides those covered by the sample exams?

    The sample exam can be found at: http://zone.ni.com/devzone/cda/epd/p/id/1662

    Thanks,

    Toby

    Download File:post-8758-1191518351.zip

  5. QUOTE(BrokenArrow @ Oct 2 2007, 09:13 AM)

    Hello all!

    I need to write, to a txt file, some of the extended ascii characters, specifically the "box drawing" type such as alt+203, alt+204, etc. How is that done in LabVIEW? I've done it with unicode (by writing FFFE 6625 for example, to get one of the characters), but as you know, the file resulting Unicode becomes useless to LV when you read it and parse for data later.

    Thanks,

    Richard

    (lv 8.2.1)

    ╩ don't think LabVIEW supports the characters you are trying to use. I attempted a similar thing a while back and never came up with a way to do it. It seems like the characters automatically get converted to something that LV can handle.

    Edit: I looked into it some more and it looks like the box characters you are trying to write (╦╔╚ etc) are actually Unicode characters and not extended ASCII. The true extended ASCII characters appear to be supported.

  6. QUOTE(Popatlal @ Oct 2 2007, 08:52 AM)

    is there any material or practical guide is available for begineer of labview...

    Welcome to Lava :D . I would highly reccomend LabVIEW for Everyone. As a beginner, this book will give you everything you need to get started and it also has some good intermediate/advanced stuff when you feel a bit more comfortable. The nice thing about a book like this is it teaches you the right way to do things before you form a lot of bad habits - and besides that, if you have any questions the author frequents this forum. Thanks Jim! :worship:

  7. QUOTE(Michael_Aivaliotis @ Oct 1 2007, 05:24 PM)

    Has anyone tried to use the new Unzip.vi? Everytime I run it, it just unzips the first file in the zip file then errors out with: Error code 2, System Exec.vi. Command was "touch 01191235142006 "filename"".

    Preview mode seems to work fine.

    Confirmed on Windows Vista as well.

  8. QUOTE(Shazlan @ Sep 30 2007, 09:55 PM)

    Dear all, I am using PXI 1422 image acquisition card to capture images from FLIR SC3000 camera. I'm currently using LabVIEW 8.2 and FLIR's Thermovision Digital Toolkit. I'm writing this LabVIEW code for one of my client. He wishes to use LabVIEW program to acquire and capture thermal images and then, use FLIR's Reseacher software to analyze these images later. So far, I am able to capture the images using LabVIEW program and everything is fine and dandy. The problem is in saving the captured images. FLIR's Reseacher can only accepts eithere .img, .seq or radiometric jpg files. The .img and .seq format is proprietary. Question 1: Does anyone knows how to save image file radiometric jpg? I've asked FLIR whether the images I've captured using IMAQ grab can be save into .img or .seq file using their Digital Toolkit and the answer is no. My discussion with FLIR can be followed by going to flir.custhelp.com and search for message # 070927-000001. There, I was told that the Reseacher reads 16-bit image file with header embedded. Question 2: Isn't jpg file saved by LabVIEW in 16-bit? Question 3: If not, please advice on how to further manipulate the images to meet the requirements.

    Has anyone here done similar things? and with great success maybe? Help if much appreciated. Shazlan

    Well, your first problem is that you are using a FLIR camera (I work for Fluke in the Thermography division so I might be biased). I don't know what tools are available in the FLIR SDK, but there should be a method you can call to capture an image with a parameter that specifies pixel type (16-bit jpg with radiometric data; 16-bit jpg with compresses png radiometric data; radiometric data only; etc). I would think you should be able to let the camera do the work for you as well. Rather than capture the image directly from the camera, call a method to capture an image on the camera (in the format that you want) and then query the camera to pull the complete image file into LabVIEW?

    The FLIR SDK manuals should explain all the options.

  9. QUOTE(allus @ Sep 20 2007, 06:54 PM)

    I need to look for a DMM which has a "send" button so the users can send the reading to the PC.

    If the user can push a "Send" button on the DMM why couldn't they push a "Send" button in your LabVIEW application that queries the DMM for its current reading? You could also implement the timed trigger that rkanders mentioned where you push a button in your LV app that waits a few seconds and then queries the DMM for data.

  10. QUOTE(ecarrig @ Sep 21 2007, 08:59 AM)

    you could try something like snagit

    I've been using SnagIt for quite a while and I could not live without it now. It is the best tool I've found for any kind of screen capture (scrolling windows, animation, movies, etc.). It is definately worth the $. :thumbup:

  11. QUOTE(geges @ Sep 13 2007, 12:16 AM)

    How Can I Programmatically Simulate a Mouse Click?

    This is my problem ...and I found this topic in NI Knowledgebase too. But I can't download the attached example file. Can you help me?

    I would like to programmatically control my mouse so that I can move to a specific control and click on it without user interaction. This would allow me to use a Mouse Up or Mouse Down Event without using Value Signaling or User Events. It will also allow me to see the behavior of the mouse on the front panel.

    The only way that I know to do this is with the User32.dll in windows. Call the SetCursorPos function and send it the horizontal and vertical values of the target you want to click, you could then call the mouse_event function in the same dll to trigger the mouse click. See example...

  12. QUOTE(Justin Goeres @ Sep 10 2007, 10:36 AM)

    But now I'm going to ###### about the UI design of that feature. When I click the "+/- QUOTE" button, it feels like I'm setting it from +QUOTE to -QUOTE. The button is labeled by its function ("what happens if I click this?") but it looks like it's labeled by its setting ("Quote Mode currently set to +QUOTE").

    On top of that, the act of Quoting is very closely related to the act of Replying, but the "+/- QUOTE" button has nothing to do with the Reply button immediately next to it! Instead, it's (totally non-obviously) coupled to the AddReply button at the bottom of the thread page, which for the record I'd never even noticed 'til now.

    I'll agree with you there. It is not very intuitive. The help files do explain it pretty well though and it's easy to use once you understand the functionality. :D

  13. The Quote +/- button works together with the "AddReply" button at the top of a thread. You can push the "Quote +/-" button on multiple entries and then push the "AddReply" button to quote all of those entries in your reply. Pushing the "Reply button will take you directly to the edit screen for a new post with that entry always quoted (regardless of the state of the Quote +/- button).

    Clear as mud?

  14. QUOTE(Neville D @ Sep 7 2007, 09:57 AM)

    I think the NI behaviour is completely consistent with normal Windoze behaviour.. if you drag and drop a file from one folder to other, it moves the file.

    Ctrl drag will copy the file over..

    That is only true if you are dragging to a location on the same partition. If you drag a file to a different partition on your hard drive (or to a different drive altogether) it will copy the file. I might agree that NI got it right if this was true in LV as well. Unfortunately I can't create two projects on separate partitions to test this out because I only have one big partition on this machine.

    Regardless, it does not seem to be consistent with Drag & Drop behavior in the rest of LV.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.