Windows 10下Qt集成OpenCV遇mingw32-make编译错误求助

Hey there, sorry to hear you're hitting a snag with mingw32-make -j 4 while setting up Qt 5.10 and OpenCV 3.4.0 on Windows. Since you didn't include the exact error output, I'll break down the most common issues folks run into with this specific setup stack—and how to fix them:

1. Mismatched MinGW Versions #

Qt 5.10 ships with a specific MinGW build (usually MinGW 5.3.0, either 32-bit or 64-bit). If you're using a separate MinGW distribution to compile OpenCV, version mismatches can trigger compiler or linker errors.

• How to fix: Use the same MinGW that came with your Qt installation. Add its bin directory (e.g., C:\Qt\5.10.0\mingw53_32\bin) to your system PATH before any other MinGW folders to ensure it's the one mingw32-make uses.

2. Botched CMake Configuration #

If your CMake setup didn't correctly point to Qt, Python, or required OpenCV modules, the generated Makefiles will fail during compilation.

• How to fix:
  • Re-run CMake and verify these critical settings:
    • Set CMAKE_BUILD_TYPE to match your Qt build (either Release or Debug—mixing them causes issues).
    • Confirm Qt5_DIR points to your Qt 5.10's CMake config folder (e.g., C:\Qt\5.10.0\mingw53_32\lib\cmake\Qt5).
    • For Python support, double-check that PYTHON3_EXECUTABLE, PYTHON3_INCLUDE_DIR, and PYTHON3_LIBRARY all point to your Python 3.6.3 installation (not a different version).
    • Enable BUILD_opencv_world—this bundles most OpenCV modules into a single library, cutting down on linker headaches.

3. Parallel Build Overloading Your System #

The -j 4 flag tells mingw32-make to use 4 concurrent threads. If your PC has less than 8GB of RAM, this can lead to out-of-memory crashes or incomplete builds.

• How to fix: Try running mingw32-make without the -j flag (single-threaded) or use a lower thread count like -j 2 to reduce resource strain.

4. Corrupted OpenCV Source Files #

Sometimes the downloaded OpenCV 3.4.0 source archive gets corrupted during download, leading to missing or broken files that break compilation.

• How to fix: Re-download the OpenCV 3.4.0 source, extract it to a fresh folder, and re-run the CMake and build steps from scratch.

5. Python Version Conflicts #

If you have multiple Python versions installed, CMake might accidentally pick the wrong one, leading to mismatched library links.

• How to fix: Temporarily remove other Python installations from your system PATH when running CMake and mingw32-make, or explicitly set the correct Python paths in CMake as noted earlier.

If you can share the exact error message (like compiler warnings, "missing file" errors, or linker failures), we can zero in on the exact problem even faster!

内容的提问来源于stack exchange，提问作者Syntechies