I set up OpenCV with Code::Blocks as per this tutorial: http://kevinhughes.ca/tutorials/opencv-install-on-windows-with-codeblocks-and-mingw/.
I've got as far as building the test program at the end successfully, but when I try to run it I get the following error: "The application was unable to start correctly (0xc00000be). Click OK to close the application."
The one thing that is certainly different from the tutorial for me is that I'm working with OpenCV 3.0.0, Code::Blocks 13.12 and the latest version of mingw (number not obvious from their website or the installer).
I found that other people had had this problem in the past, e.g. The application was unable to start correctly (0xc00000be)
However, it has been a long time since that thread was active, I'm working with newer versions of the software and, most importantly, the answer given there didn't help (I compiled openCV with the same mingw that I'm using with Code::Blocks), so I thought this question would be worth asking (sorry if it's not, I'm rather new to this).
Could anybody suggest a fix that might get rid of this error?
A few other things to note:
I didn't download the Code::Blocks that came with mingw as I was advised against it, I got them separately as per the tutorial
I have installed OpenCV and Code::Blocks on my E: drive, though mingw is still on my C:.
I have added both mingw and OpenCV to my PATH variable, and mingw to my Path variable
I have linked the libraries in both the main compiler settings for Code::Blocks and the Build Options of my project.
I'm working from a 64-bit Windows 7 Laptop, but I have done everything 32-bit style as advised in the tutorial.
Thanks in advance!
So, one solution I've found to this problem is to copy all the .dll files from "...\opencv\build\x86\mingw\bin" to the project file (build directory).
Having done that, the project will run fine. This still doesn't (directly) explain why it wont run without those files (at least, several sources including the tutorial imply it should), and so isn't really optimal. I will continue to look for a solution for this so I don't have to clutter every opencv project I make with the .dlls.
However, for anyone simply wanting to get stuck into coding who is having this problem, I recommend that solution.
Related
Firstly, thank you in advance for at least reading this question.
Secondly, feel free to call me an idiot if this is a "no brainer"
Now down to the issue at hand...
I am running Ubuntu 16.04 (64bit) with QT Creator as my IDE and on the whole it works great. I can cross compile for the Raspberry Pi3 (running raspbian) and also compile plugins for X-Plane.
Everything was going great until I needed to implement MySql into my project. That's when it all went south.
I can not use the Qt library for this as X-Plane does not like anything Qt related in its plugins. But that is fine, I have connector-c++ working on that machine and it connects to my MySql server and does everything it's supposed to.
I think I have read just about every forum and watched every youtube video on this but can not get Qt Creator to compile with connector for the raspberry pi.
I wrote/compiled a test app locally (with g++) on the pi and it works great. However for some reason (government plot maybe?) it refuses to link the libraries when compiling under Qt Creator.
The error I get when compiling in QtCreator is:
undefined reference to `get_driver_instance'
Currently,
I have followed the instructions for setting up a cross compiler here:
https://wiki.qt.io/RaspberryPi2EGLFS
Downloaded the connector code from here:
git clone https://github.com/mysql/mysql-connector-cpp
I have followed the instructions for compiling connector-c++ here:
https://dev.mysql.com/doc/connector-cpp/en/connector-cpp-installation-source-unix.html
And I have banged my head 1000x on a brick wall here:
(picture of blood stained wall not available)
Also, I have tried compiling connector on my Ubuntu pc with paths to the pi's sysroot etc. but i think it made a 64bit version which causes Qt Creator to have a dummy spit about "unrecognized format".
I am hoping that this is just me having a "senior moment" and that someone can point me in the right direction.
To save putting up irrelevant logs/code I will wait until someone asks for it and then copy/paste what they need.
Again, thanks in advance and know that I will be eternally grateful for any light on this matter (preferably before I go even insanerer)
Well after 2 weeks of pain, I have solved this problem. Here is what I think is the issue:
Firstly, I found that mysqlcppconn is not needed (at least for me) and used the mysql libraries in /usr/lib
Secondly, the symlinks to libmysqlclient.so.6.0.22 were by default:
libmysqlclient.so.6 -> libmysqlclient.so.6.0.22
The compiler was looking for libmysqlclient.so which didn't exist. So I created the link (libmysqlclient.so -> libmysqlclient.so.6) and it worked!!
Thanks to all who helped.
I have OSX Yosemite 10.10.5, QtCreator 4.1.0, Qt 5.7.0 and Xcode 7.2.1.
If I write any native C++ program in QtCreator (console, gui - anything) I am unable to debug that program from QtCreator. Breakpoints get ignored, the program executes without pause and runs to completion no matter what I try.
I have tried the Xcode's LLDb, system GDB and Homebrew GDB as debuggers. the versions of GDB fail to start. Homebrew GDB has been codesigned.
This all used to (a few year back) just work beautifully and I am at a loss as to what might have changed.
Curiously, if I generate an exception inside the program - like accessing through a null pointer, the debugger shows me where this happens in just the way I would expect so, presumably, the debugger is running but simply treating me with complete disdain.
I know there are many versions of this question but none seem to address the problem adequately or offer workable solutions, or they apply to much older versions of the products.
Any suggestions?
This version of qtCreator (4.1.0) saw the return of the warning about having the build directory at the same level as the project directory. In Windows, this is done automatically. My mac installation does not get it right. (I may have messed a while back - cannot remember).
So - if the build directory is inside the project directory, debugging fails. Move the build directory up to the same level as the project directory and everything seems to work fine.
You can set the default build directory in the preferences/Build & Run dialogue. The path should start with ../
You will need to think about the folder structure before setting up the project unless you want build folders appearing in awkward places. So, a project folder structure like this is OK:
But one like this is not:
If this was already obvious to you, great. I have been messing with this for ages.
Many thanks to those who replied and anyone else who took the time to read the question. What helped e find it was the suggestion by #AlexanderVX that his setup was the same as mine but his worked. The only bit of my setup he could not see in the screenshots was the start of the build path. So that was the clue.
Following this tutorial, I've created my positive samples but need to merge them now, using mergevec. I downloaded the mergevec.exe binary file provided and got the two required dlls cxcore100.dll and highgui100.dll. However, when I run it like so:
mergevec samples.txt samples.vec
it gives me the error:
ERROR: Input file <filename> does not exist or not readable.
What should I do to fix this? My vec files are fine, because I was able to view them using the opencv_createsamples utility. So I know they're fine.
I ran into this issue as well and made a python utility that combines .vec files to avoid installing openCV again: https://github.com/wulfebw/mergevec
Hope it helps.
I was in the same situation as you few days ago so i'll try to help you out =P.
First of all i am interested to know where you found the mergevec.exe, because it isn't available to download anymore if i recall. I ask because i think that you might have downloaded an exe compiled on Linux. I had the same error when i tried to execute the mergevec, i had compiled on Ubuntu, on my Windows computer. Even with the 2 dlls.
As i explained it here , The mergevec from Naotoshi was originally made for Unix distribution, so under windows you will encounter a lot of problems. I first tried to do it on windows but i abandonned because it was too complicated and even with a full Cygwin installation i could not compile the mergevec.cpp.
What i advise you to do is to create yourself an Ubuntu virtual machine and to compile your own mergevec.exe. Once done you can copy your vec files in your virtual machine and merge them under Ubuntu. After this you can bring back the merged vec file to Windows if you want (It is better to not run the training under a VM: less computation power).
Here is what i did:
Installed VMware player with Ubuntu 14.0.4 (although i don't recommend it, still some bugs)
Installed OpenCV from the sources with TBB. There are a lot of tutorials, but the only one that worked for me is the OPenCV website, they are the more accurate on the libraries you mandatorily need.
I downloaded the GitHub classifier training and then I followed the commands given and it worked well.
If you encounter any library problem or OpenCV Installation issue don't hesitate i went through them too.
I have successfully built OpenCV and I have come to the point where code::blocks recognizes the OpenCV syntax and successfully builds the program. However when I run I get this error
"The procedure entry point_gxx_personality_v0 could not be located in the dynamic link library libstdc++-6.dll."
I am using Code::Blocks v 12.11 and minGW (the most recent, I can't remember the version number) along with OpenCV-2.4.8 Has anyone ever seen this problem? Help if you can, anything right now might work since I am so close.
https://stackoverflow.com/questions/3895879/installing-opencv-on-windowsw32-to-be-used-with-code-blocks?rq=1
Using this link to the other question I was able to locate a hint for my problem. I had everything correct and the hint to change the build to a GUI Application solved the issue. The programs build and then run as they should.
I'm new to this community and this is my first question. I know many people had had this problem before, but, it seems as if people solved it their own ways and I tried all the solutions I could find in google (and here, respectively), but none of them seem to work for me like it was supposed to. And just to make it clear, I use 32-bit Windows XP. Some versions that may work in 64-bit or newer operating systems may not work on mine.
So I'm very new to C and C++, I started by installing the Eclipse CDT, the moment I found out it did not have a compiler was when I installed MinGW which was bundled with a C compiler, a C++ compiler, and the MinGW Developer's Toolkit (which included MinSys). So after that, I started by creating a new empty C++ project called 'HyScript'. And did these:
I set the environment variables as:
C:\Program Files\Java\jre7\bin;E:\Mark\MinGW\msys\1.0\bin;E:\Mark\MinGW\bin
Then, I restarted eclipse, deleted the old projects, and created a new one with the same name (deleted the old folders) and chose the toolchair, "Cross GCC".
Now I haven't even started writing code, but there are already 4 errors:
Here's the link to the dropbox file
The last thing I did was open up the CMD before going here, and this happened: Here's another link to the dropbox file
I don't understand anymore... What did I do wrong? I set the environment variables right, I supposed. And, after doing some research, I still can't find a proper solution for me. It's been two days now, and this error still hasn't gone away.
You chose wrong toolchair "Cross GCC". Please select "MinGW"
The latest version of G++ can be found here. Please do NOT install it in the default directory but use this:
C:\MinGW
If you still have problems, uninstall everything you did, and watch these video instructions.