Revision history [back]

I have now gone for a workaround as described below.

Seemingly no one can explain what “try to: cmake –OPENCV_ENABLE_ALLOCATOR_STATS=OFF” means even though it is the apparent solution to my issue. I have spent 2 days trying to work it out for myself but with no luck.

I was trying to follow the implementation here. https://medium.com/@sourabhjigjinni/install-opencv-4-0-0-for-c-windows-7-10-code-blocks-tdm-gcc-64-dff65addf162, so my workaround is to follow the excellent guide there but to use the versions of software as given there. (Open CV 4.1.0, cmake 3.13.0, tdm-gcc 5.1.0 and CodeBlocks 20.03). With these versions I no longer get the error “error long no such directory”.

I am left with a dilemma. I recently identified a fault with a longstanding image processing method and I provided a solution in Matlab (Ruggedised Xiao Method https://github.com/hangong/Xiao06_color_transfer. ) It was my intention to provide a version of this in openCV but given my experiences here I am having second thoughts.

I have now gone for a workaround as described below.

Seemingly no one can explain My original problem was that i did not understand what “try to: cmake –OPENCV_ENABLE_ALLOCATOR_STATS=OFF” means even though it is the apparent solution to my issue. I have spent 2 days trying to work it out for myself but with no luck.

“cmake - OPENCV_ENABLE_ALLOCATOR_STATS=OFF”means”.

The comment below answers this. (Thanks to 'bhilareakshay1')

The instructions that I was trying to follow the implementation are here. https://medium.com/@sourabhjigjinni/install-opencv-4-0-0-for-c-windows-7-10-code-blocks-tdm-gcc-64-dff65addf162, so my workaround is to follow the excellent guide there
https://medium.com/@sourabhjigjinni/install-opencv-4-0-0-for-c-windows-7-10-code-blocks-tdm-gcc-64-dff65addf162
The instructions included at step 5.3 the need to tick and untick various options. I now realise that as part of this step I should additionally untick ‘OPENCV_ENABLE_ALLOCATOR_STATS .

This answers the question that I originally asked but to use the versions of software as given there. (Open CV 4.1.0, cmake 3.13.0, tdm-gcc 5.1.0 and CodeBlocks 20.03). With these versions I no longer get the error “error long no such directory”.

I am left it still leaves me with a dilemma. I recently identified further problem. When I implement the fix I still get an error.

“C:\TDM-GCC-64\lib\gcc\x86_64-w64-mingw32\9.2.0\include\c++\math.h |91|error: ‘std::_hypot’ has not been declared| modules\python3\CMakeFiles\opencv_python3.dir\build.make|82|recipe for target ‘modules/python3/CMakeFiles/ opencv_python3.dir/__/src2/cv2.cpp.obj’ failed| …………. etc”

I do not intend to raise this as a fault with a longstanding image processing method and I provided a solution in Matlab (Ruggedised Xiao Method https://github.com/hangong/Xiao06_color_transfer. ) It was my intention to provide a version of this in openCV but given my experiences new issue because I can implement OpenCV 4.1.0 rather than Open CV 4.3.0 successfully by following the instructions that I previously referenced.

For the record, therefore using the instructions here I am having second thoughts.

My original problem was that i did not understand what “cmake - OPENCV_ENABLE_ALLOCATOR_STATS=OFF”means”.

The comment below answers this. (Thanks to 'bhilareakshay1')

The instructions that I was trying to follow are here.
https://medium.com/@sourabhjigjinni/install-opencv-4-0-0-for-c-windows-7-10-code-blocks-tdm-gcc-64-dff65addf162
The instructions included at step 5.3 the need to tick and untick various options. I now realise that as part of this step I should additionally untick ‘OPENCV_ENABLE_ALLOCATOR_STATS .

This answers the question that I originally asked but it still leaves me with a further problem. When I implement the fix I still get an error.

“C:\TDM-GCC-64\lib\gcc\x86_64-w64-mingw32\9.2.0\include\c++\math.h |91|error: ‘std::_hypot’ has not been declared| modules\python3\CMakeFiles\opencv_python3.dir\build.make|82|recipe for target ‘modules/python3/CMakeFiles/ opencv_python3.dir/__/src2/cv2.cpp.obj’ failed| …………. etc”

I do not intend to raise this as a new issue because I can implement OpenCV 4.1.0 rather than Open CV 4.3.0 successfully by following the instructions that I previously referenced.

For the record, therefore using the instructions here

https://medium.com/@sourabhjigjinni/install-opencv-4-0-0-for-c-windows-7-10-code-blocks-tdm-gcc-64-dff65addf162

I can install OpenCV using the following software versions.

https://cmake.org/files/v3.13/cmake-3.13.0-win64-x64.msi

https://sourceforge.net/projects/codeblocks/files/Binaries/20.03/Windows/codeblocks-20.03-setup.exe/download?use_mirror=autoselect

https://sourceforge.net/projects/tdm-gcc/files/TDM-GCC%20Installer/tdm64-gcc-5.1.0-2.exe/download

https://github.com/opencv/opencv/archive/4.1.0.zip

Now that I fell feel that I can reliably install OpenCV under windows, I have started again to post C++/OpenCV code upon GitHub for colour transfer methods as follows.

https://github.com/TJCoding/Ruggedised-Image-Colour-Transfer https://github.com/TJCoding/Pitie_Image_Colour_Transfer

My original problem was that i did not understand what “cmake - -D OPENCV_ENABLE_ALLOCATOR_STATS=OFF”means”.

The comment below answers this. (Thanks to 'bhilareakshay1')

The instructions that I was trying to follow are here.
https://medium.com/@sourabhjigjinni/install-opencv-4-0-0-for-c-windows-7-10-code-blocks-tdm-gcc-64-dff65addf162
The instructions included at step 5.3 the need to tick and untick various options. I now realise that as part of this step I should additionally untick ‘OPENCV_ENABLE_ALLOCATOR_STATS .

This answers the question that I originally asked but it still leaves me with a further problem. When I implement the fix I still get an error.

“C:\TDM-GCC-64\lib\gcc\x86_64-w64-mingw32\9.2.0\include\c++\math.h |91|error: ‘std::_hypot’ has not been declared| modules\python3\CMakeFiles\opencv_python3.dir\build.make|82|recipe for target ‘modules/python3/CMakeFiles/ opencv_python3.dir/__/src2/cv2.cpp.obj’ failed| …………. etc”

I do not intend to raise this as a new issue because I can implement OpenCV 4.1.0 rather than Open CV 4.3.0 successfully by following the instructions that I previously referenced.

For the record, therefore using the instructions here

https://medium.com/@sourabhjigjinni/install-opencv-4-0-0-for-c-windows-7-10-code-blocks-tdm-gcc-64-dff65addf162

I can install OpenCV using the following software versions.

https://cmake.org/files/v3.13/cmake-3.13.0-win64-x64.msi

https://sourceforge.net/projects/codeblocks/files/Binaries/20.03/Windows/codeblocks-20.03-setup.exe/download?use_mirror=autoselect

https://sourceforge.net/projects/tdm-gcc/files/TDM-GCC%20Installer/tdm64-gcc-5.1.0-2.exe/download

https://github.com/opencv/opencv/archive/4.1.0.zip

Now that I feel that I can reliably install OpenCV under windows, I have started again to post C++/OpenCV code upon GitHub for colour transfer methods as follows.

https://github.com/TJCoding/Ruggedised-Image-Colour-Transfer https://github.com/TJCoding/Pitie_Image_Colour_Transfer