http s://cloud.189.cn/t/QfArq2i6n6j2 (访问码:ttv0)参考了https://bbs.huaweicloud.com/forum/thread-53302-1-1.html这个是编译好的。是一个tar包，不要在windows下解压，直接拖到hilens里边解压。压缩包内的文件如下： 在hilens中需要替换以下文件。请替换之前做好备份，搞坏不负责。 /usr/lib64/python3.7/site-packages/cv2.cpython-37m-aarch64-linux-gnu.so 替换为 上图python3文件夹中的 文件。注意，需要改名。将压缩包内的.so后缀的文件，扔到 /usr/lib64/中。/home/hilens/skillframework/lib下libopencv_world.so，替换成新产生的libopencv_world.so

Error running link command: No such file or directory3rdparty/quirc/CMakeFiles/quirc.dir/build.make:146: recipe for target '3rdparty/lib/libquirc.a' failedmake[2]: *** [3rdparty/lib/libquirc.a] Error 2CMakeFiles/Makefile2:640: recipe for target '3rdparty/quirc/CMakeFiles/quirc.dir/all' failedmake[1]: *** [3rdparty/quirc/CMakeFiles/quirc.dir/all] Error 2make[1]: *** 正在等待未完成的任务....此外：cmake 版本为3.5.1

【转载华为云社区】人脸检测技术在我们生活中经常被用到，人脸面部识别、年龄以及表情的识别都是以人脸为基础。接下来我分享一下最近整理的开源人脸检测工具，由于篇幅较长，所以打算将此作为一个专栏，希望能给AI实战营的小伙伴们提供帮助！目前常用的人脸检测工具有：OpenCV Haar级联分类器、MTCNN方式、dlib方式...（后续会继续补充），首先介绍一下OpenCV.....什么是OpenCV？OpenCV（Open Source Computer Vision Library）是一个开源的计算机视觉库，它提供了很多函数，这些函数非常高效地实现了计算机视觉算法（最基本的滤波到高级的物体检测皆有涵盖）。OpenCV 使用 C/C++ 开发，同时也提供了 Python、Java、MATLAB 等其他语言的接口。OpenCV 是跨平台的，可以在  Windows、Linux、Mac OS、Android、iOS 等操作系统上运行。OpenCV 的应用领域非常广泛，包括图像拼接、图像降噪、产品质检、人机交互、人脸识别、动作识别、动作跟踪、无人驾驶等。(以上源于百度搜索)什么是级联分类器？大致知道OpenCV的作用后，来看下Haar特征+Adaboost级联分类器，Haar特征算子就是下图的所有黑色矩阵的像素值减去所有白色矩阵像素值的和。Haar特征可以在检测窗口中由放大+平移产生一系列子特征。可以简单理解为算子在图片上横扫过去生成多个框检索是否有人脸的存在准备工作打开ModelArts控制台选择notebook：点击new，选择Tensorflow1.13.1框架。如图：注：Modelarts中安装好了OpenCV供我们使用，无需再次安装。将测试图片先上传到notebook中，如图，点击Upload上传自己的本地图片：代码实战将haarcascade_frontalface_alt.xml用path变量存储起来。接下来配置分类器，读取图片后进行灰度化处理，处理完将图片传入检测器中检测。haarcascade_frontalface_alt.xml文件可以理解为模型，存储了6061个特征，38个阶段，算子的位置大小形状，直接使用即可。下图给出detectMultiScale检测函数中的参数介绍：第六行代码中detectMultiScale方法中和第一种方法中介绍的一样，gray就是image，1.3代表scaleFactor,5是minNeighbors。最后输出图片结果：解释下最后一句from PIL import Image，PIL：Python Imaging Library，已经是Python平台事实上的图像处理标准库。由于我们对图片进行灰度处理，需要从numpy数组格式转换回image格式，故使用Image.fromarray(img)。文中所述的完整代码如下，欢迎大家参考学习！path="/home/ma-user/anaconda3/envs/TensorFlow-1.13.1/lib/python3.6/site-packages/cv2/data/haarcascade_frontalface_alt.xml" import cv2 face_cascade = cv.CascadeClassifier(path) #修改为自己的图片文件名 image_path = "./img_nini.jpeg" img = cv.imread(image_path) gray = cv.cvtColor(img,cv.COLOR_BGR2GRAY) faces = face_cascade.detectMultiScale(gray,1.3,5) for(x,y,w,h) in faces:     cv.rectangle(img,(x,y),(x+w,y+h),(0,255,0),2) img = cv.cvtColor(img,cv.COLOR_BGR2RGB) from PIL import Image Image.fromarray(img)

请教：有没有atlas500安装opencv的教程？

编译opencv4.1.2，4.2.0都会出现这个错误，已经成功编译了x264，gflags，ffmpeg都可以输入命令:cmake -DCMAKE_TOOLCHAIN_FILE=/home/ubuntu/huawei/Euler.cmake -DCMAKE_INSTALL_PREFIX=/home/ubuntu/huawei/huaweilib/opencv -DWITH_WEBP=OFF -DBUILD_opencv_world=ON ..make 

新手求助，如何在atlas500上编译安装opencv?

遇到同样的问题，在windows上可以正常打开视频，在其他环境下也可以用opencv正常打开视频，也安装了ffmpeg代码如下：结果如下：

请问ATLAS 200DK目标跟踪算法库怎么添加呢？程序中创建KCF对象 诸如Ptr<TrackerKCF> tracker=TrackerKCF::creat();时报错，头文件中已#include<opencv2/tracking.hpp> 和#include<opencv2/tracking/tracker.hpp>  build时都不通过。问题求助！！

交叉编译BGSLibrary的时候，因为要添加OpenCV的依赖，于是传入了DDK中OpenCV的路径，但是在执行make时，报错找不到OpenCV的库方法传入路径如下：set(OpenCV_VERSION 4.1.0)set(OpenCV_LIBS /opt/atlas200/ddk/RC/host-aarch64_ubuntu16.04/lib/libopencv_world.so.4.1)set(OpenCV_INCLUDE_DIRS /opt/atlas200/ddk/include/third_party/opencv/include/opencv2/)make报错如下：[ 99%] Building CXX object CMakeFiles/bgslibrary.dir/src/VideoCapture.cpp.o[100%] Linking CXX executable bgslibraryCMakeFiles/bgslibrary.dir/src/FrameProcessor.cpp.o: In function `bgslibrary::FrameProcessor::process(cv::Mat const&)':/home/***********/code/bgslibrary-master/src/FrameProcessor.cpp:184: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'CMakeFiles/bgslibrary.dir/src/PreProcessor.cpp.o: In function `cv::Mat::clone() const':/usr/include/opencv2/core/mat.hpp:335: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'CMakeFiles/bgslibrary.dir/src/PreProcessor.cpp.o: In function `bgslibrary::PreProcessor::process(cv::Mat const&, cv::Mat&)':/home/***********/code/bgslibrary-master/src/PreProcessor.cpp:33: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'/home/***********/code/bgslibrary-master/src/PreProcessor.cpp:37: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'/home/***********/code/bgslibrary-master/src/PreProcessor.cpp:51: undefined reference to `cv::_InputArray::_InputArray(cv::Mat const&)'/usr/include/opencv2/core/core.hpp:4733: undefined reference to `cv::CommandLineParser::getString(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'/usr/include/opencv2/core/core.hpp:4734: undefined reference to `std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > cv::CommandLineParser::analyzeValue<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)'CMakeFiles/bgslibrary.dir/src/VideoAnalysis.cpp.o: In function `int cv::CommandLineParser::get<int>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)':/usr/include/opencv2/core/core.hpp:4733: undefined reference to `cv::CommandLineParser::getString(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'/usr/include/opencv2/core/core.hpp:4734: undefined reference to `int cv::CommandLineParser::analyzeValue<int>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)'CMakeFiles/bgslibrary.dir/src/VideoAnalysis.cpp.o: In function `std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > cv::CommandLineParser::get<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)':/usr/include/opencv2/core/core.hpp:4733: undefined reference to `cv::CommandLineParser::getString(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'/usr/include/opencv2/core/core.hpp:4734: undefined reference to `std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > cv::CommandLineParser::analyzeValue<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)'CMakeFiles/bgslibrary.dir/src/VideoAnalysis.cpp.o: In function `int cv::CommandLineParser::get<int>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)':/usr/include/opencv2/core/core.hpp:4733: undefined reference to `cv::CommandLineParser::getString(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'/usr/include/opencv2/core/core.hpp:4734: undefined reference to `int cv::CommandLineParser::analyzeValue<int>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)'CMakeFiles/bgslibrary.dir/src/VideoCapture.cpp.o: In function `bgslibrary::VC_ROI::VideoCapture_on_mouse(int, int, int, int, void*)':/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:60: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:63: undefined reference to `cv::rectangle(cv::Mat&, cv::Point_<int>, cv::Point_<int>, cv::Scalar_<double> const&, int, int, int)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:64: undefined reference to `cv::_InputArray::_InputArray(cv::Mat const&)'CMakeFiles/bgslibrary.dir/src/VideoCapture.cpp.o: In function `bgslibrary::VideoCapture::setUpCamera()':/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:100: undefined reference to `cv::VideoCapture::open(int)'CMakeFiles/bgslibrary.dir/src/VideoCapture.cpp.o: In function `bgslibrary::VideoCapture::setUpVideo()':/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:116: undefined reference to `cv::VideoCapture::open(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'CMakeFiles/bgslibrary.dir/src/VideoCapture.cpp.o: In function `bgslibrary::VideoCapture::start()':/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:147: undefined reference to `cv::VideoCapture::get(int)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:164: undefined reference to `cv::_InputArray::_InputArray(cv::Mat const&)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:164: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:224: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:243: undefined reference to `cv::_InputArray::_InputArray(cv::Mat const&)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:240: undefined reference to `cv::putText(cv::Mat&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, cv::Point_<int>, int, double, cv::Scalar_<double>, int, int, bool)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:183: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat&)'/home/***********/code/bgslibrary-master/src/VideoCapture.cpp:187: undefined reference to `cv::_InputArray::_InputArray(cv::Mat const&)'libbgslibrary_core.so: undefined reference to `cv::findContours(cv::_OutputArray const&, cv::_OutputArray const&, int, int, cv::Point_<int>)'libbgslibrary_core.so: undefined reference to `cv::BackgroundSubtractorMOG2::BackgroundSubtractorMOG2()'libbgslibrary_core.so: undefined reference to `cv::BackgroundSubtractor::~BackgroundSubtractor()'libbgslibrary_core.so: undefined reference to `cv::drawContours(cv::_OutputArray const&, cv::_InputArray const&, int, cv::Scalar_<double> const&, int, int, cv::_InputArray const&, int, cv::Point_<int>)'libbgslibrary_core.so: undefined reference to `cv::accumulateProduct(cv::_InputArray const&, cv::_InputArray const&, cv::_OutputArray const&, cv::_InputArray const&)'libbgslibrary_core.so: undefined reference to `cv::DescriptorExtractor::~DescriptorExtractor()'libbgslibrary_core.so: undefined reference to `cv::BackgroundSubtractorMOG2::~BackgroundSubtractorMOG2()'libbgslibrary_core.so: undefined reference to `cv::initModule_video()'libbgslibrary_core.so: undefined reference to `cv::Algorithm::set(char const*, double)'libbgslibrary_core.so: undefined reference to `cv::cvarrToMat(void const*, bool, bool, int)'libbgslibrary_core.so: undefined reference to `cv::Algorithm::read(cv::FileNode const&)'libbgslibrary_core.so: undefined reference to `cv::Algorithm::write(cv::FileStorage&) const'libbgslibrary_core.so: undefined reference to `vtable for cv::_InputArray'libbgslibrary_core.so: undefined reference to `cv::BackgroundSubtractorMOG2::operator()(cv::_InputArray const&, cv::_OutputArray const&, double)'libbgslibrary_core.so: undefined reference to `cv::BackgroundSubtractorMOG::~BackgroundSubtractorMOG()'libbgslibrary_core.so: undefined reference to `cv::Algorithm::set(char const*, int)'libbgslibrary_core.so: undefined reference to `cv::Mat::operator _IplImage() const'libbgslibrary_core.so: undefined reference to `cv::split(cv::Mat const&, std::vector<cv::Mat, std::allocator<cv::Mat> >&)'libbgslibrary_core.so: undefined reference to `cv::_OutputArray::_OutputArray(cv::Mat const&)'libbgslibrary_core.so: undefined reference to `typeinfo for cv::FeatureDetector'libbgslibrary_core.so: undefined reference to `cv::FeatureDetector::empty() const'libbgslibrary_core.so: undefined reference to `cvCalcOpticalFlowPyrLK'libbgslibrary_core.so: undefined reference to `cv::_InputArray::_InputArray(cv::MatExpr const&)'libbgslibrary_core.so: undefined reference to `cv::DescriptorExtractor::empty() const'libbgslibrary_core.so: undefined reference to `cv::Algorithm::_create(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)'libbgslibrary_core.so: undefined reference to `cv::BackgroundSubtractorMOG::BackgroundSubtractorMOG()'libbgslibrary_core.so: undefined reference to `cv::FeatureDetector::~FeatureDetector()'libbgslibrary_core.so: undefined reference to `typeinfo for cv::DescriptorExtractor'libbgslibrary_core.so: undefined reference to `cv::BackgroundSubtractor::operator()(cv::_InputArray const&, cv::_OutputArray const&, double)'collect2: error: ld returned 1 exit statusCMakeFiles/bgslibrary.dir/build.make:146: recipe for target 'bgslibrary' failedmake[2]: *** [bgslibrary] Error 1CMakeFiles/Makefile2:109: recipe for target 'CMakeFiles/bgslibrary.dir/all' failedmake[1]: *** [CMakeFiles/bgslibrary.dir/all] Error 2Makefile:129: recipe for target 'all' failedmake: *** [all] Error 2

由于需要用到opencv的一些功能，在Mindstudio中需要配置一下参考了sample-facialrecogntion的例程face_feature_mask.cpp，引用的opencv的头文件 #include“opencv2/opencv.hpp”,在CMakeLists.txt中$ENV{DDK_PATH}/include/third_party/opencv/include    target_link_libraries(opencv_world)，在main.cpp中添加opencv的头文件。编译没问题，但是添加 Mat scrimage 函数后编译报错，error:undefined reference to 'cv::fastFree(void*)'请问大神，这是什么原因，是Mindstudio那配置有问题吗？感谢

如图，在执行下面步骤时出现问题问题如下：两次执行都是如此，求解答问题出在哪。

如下，目前需要实现圆形指针表和数字表的检测读取功能，读出表的数字，上位机上的办法是用OPENCV的库进行处理，请问Atlas200DK的基本思路是什么？是在编译的Ubuntu环境下安装OPENCV然后再Mindstuio里调用OPENCV库函数对图片进行处理吗？这种情况下Atlas200DK 开发板上还需要安装OPENCV吗？请大神给个思路，或者样例。

请问各位，我在camera_ap_to_ros中添加了一部分图像处理代码后，编译时报错如下，找不到opencv2。/home/yue/eclipse-workspace/cameraros1910/src/camera_to_ros.h:18:30: fatal error: opencv2/opencv.hpp: No such file or directory如图includes里添加了路径后依然找不到请问是为什么呢？谢谢

在运行.py文件时提示找不到opencv。但是已经装了opencv了。系统python环境下import cv2。也没有问题。但是运行.py出了问题。

atlas中可以安装opencv4.2 版本么？交叉编译源码opencv4.2出现问题：#error "OpenCV 4.x+ requires enabled C++11 support"，请问这个是不是跟交叉编译链的版本不一样有关啊，这样写aarch64-linux-gnu-g++ `pkg-config --cflags --libs opencv4` test.cpp -lpthread -lrt -o test -std=c++11出现错误：/tmp/ccYBnaRi.o：在函数‘main’中：test.cpp:(.text+0x90)：对‘cv::imread(std::string const&, int)’未定义的引用test.cpp:(.text+0xc8)：对‘cv::namedWindow(std::string const&, int)’未定义的引用test.cpp:(.text+0x10c)：对‘cv::imshow(std::string const&, cv::_InputArray const&)’未定义的引用collect2: error: ld returned 1 exit status