如何通过共享内存将 cv::Mat 发送到 python?

ant*_*nti 4 c++ python opencv shared-memory

我有一个 C++ 应用程序,它通过共享内存将数据发送到 python 函数。这ctypes在 Python 中非常有效,例如双精度和浮点数。现在,我需要cv::Mat在函数中添加一个。

我的代码目前是:

//H

#include <iostream>
#include <opencv2\core.hpp>
#include <opencv2\highgui.hpp>


struct TransferData
{
   double score;
   float other;  
   int num;
   int w;
   int h;
   int channels;
   uchar* data;

};

#define C_OFF 1000
void fill(TransferData* data, int run, uchar* frame, int w, int h, int channels)
{
   data->score = C_OFF + 1.0;
   data->other = C_OFF + 2.0;
   data->num = C_OFF + 3;   
   data->w = w;
   data->h = h;
   data->channels = channels;
   data->data = frame;
}
Run Code Online (Sandbox Code Playgroud)

//.cpp

namespace py = pybind11;

using namespace boost::interprocess;

void main()
{

    //python setup
    Py_SetProgramName(L"PYTHON");
    py::scoped_interpreter guard{};
    py::module py_test = py::module::import("Transfer_py");


    // Create Data
    windows_shared_memory shmem(create_only, "TransferDataSHMEM",
        read_write, sizeof(TransferData));

    mapped_region region(shmem, read_write);
    std::memset(region.get_address(), 0, sizeof(TransferData));

    TransferData* data = reinterpret_cast<TransferData*>(region.get_address());


    //loop
    for (int i = 0; i < 10; i++)
    {
        int64 t0 = cv::getTickCount();

        std::cout << "C++ Program - Filling Data" << std::endl;

        cv::Mat frame = cv::imread("input.jpg");

        fill(data, i, frame.data, frame.cols, frame.rows, frame.channels());

        //run the python function   

        //process
        py::object result = py_test.attr("datathrough")();


        int64 t1 = cv::getTickCount();
        double secs = (t1 - t0) / cv::getTickFrequency();

        std::cout << "took " << secs * 1000 << " ms" << std::endl;
    }

    std::cin.get();
}
Run Code Online (Sandbox Code Playgroud)

//Python //传输数据类

import ctypes


    class TransferData(ctypes.Structure):
_fields_ = [
    ('score', ctypes.c_double),
    ('other', ctypes.c_float),       
    ('num', ctypes.c_int),
    ('w', ctypes.c_int),
    ('h', ctypes.c_int),
    ('frame', ctypes.c_void_p),
    ('channels', ctypes.c_int)  
]


    PY_OFF = 2000

    def fill(data):
        data.score = PY_OFF + 1.0
        data.other = PY_OFF + 2.0
        data.num = PY_OFF + 3
Run Code Online (Sandbox Code Playgroud)

//主要的Python函数

import TransferData
import sys
import mmap
import ctypes




def datathrough():
    shmem = mmap.mmap(-1, ctypes.sizeof(TransferData.TransferData), "TransferDataSHMEM")
    data = TransferData.TransferData.from_buffer(shmem)
    print('Python Program - Getting Data')   
    print('Python Program - Filling Data')
    TransferData.fill(data)
Run Code Online (Sandbox Code Playgroud)

如何将cv::Mat帧数据添加到 Python 端?我将它作为uchar*来自 c++ 的a 发送,据我所知,我需要它是一个numpy数组才能cv2.Mat在 Python 中获取 a 。从“宽度、高度、通道、frameData”到 opencv python 的正确方法是什么cv2.Mat

我使用共享内存是因为速度是一个因素,我已经使用 Python API 方法进行了测试,但它对于我的需求来说太慢了。

Dan*_*šek 5

总体思路(在 OpenCV Python 绑定中使用)是创建一个ndarrayMat对象共享其数据缓冲区的 numpy ,并将其传递给 Python 函数。

注意:此时,我将示例仅限于连续矩阵。

我们可以利用pybind11::array课堂。

  • 我们需要确定适合dtypenumpy 数组使用。这是一个简单的 1 对 1 映射,我们可以使用switch

    py::dtype determine_np_dtype(int depth)
    {
        switch (depth) {
        case CV_8U: return py::dtype::of<uint8_t>();
        case CV_8S: return py::dtype::of<int8_t>();
        case CV_16U: return py::dtype::of<uint16_t>();
        case CV_16S: return py::dtype::of<int16_t>();
        case CV_32S: return py::dtype::of<int32_t>();
        case CV_32F: return py::dtype::of<float>();
        case CV_64F: return py::dtype::of<double>();
        default:
            throw std::invalid_argument("Unsupported data type.");
        }
    }
    
    Run Code Online (Sandbox Code Playgroud)
  • 确定 numpy 数组的形状。为了使其行为类似于 OpenCV,让我们将 1-channel Mats映射到 2D numpy 数组,并将多通道Mats映射到 3D numpy 数组。

    std::vector<std::size_t> determine_shape(cv::Mat& m)
    {
        if (m.channels() == 1) {
            return {
                static_cast<size_t>(m.rows)
                , static_cast<size_t>(m.cols)
            };
        }
    
        return {
            static_cast<size_t>(m.rows)
            , static_cast<size_t>(m.cols)
            , static_cast<size_t>(m.channels())
        };
    }
    
    Run Code Online (Sandbox Code Playgroud)
  • 提供将共享缓冲区的生命周期扩展到 numpy 数组的生命周期的方法。我们可以pybind11::capsule围绕源代码创建一个浅拷贝Mat——由于对象的实现方式,这在所需的时间内有效地增加了它的引用计数。

    py::capsule make_capsule(cv::Mat& m)
    {
        return py::capsule(new cv::Mat(m)
            , [](void *v) { delete reinterpret_cast<cv::Mat*>(v); }
            );
    }
    
    Run Code Online (Sandbox Code Playgroud)

现在,我们可以执行转换了。

py::array mat_to_nparray(cv::Mat& m)
{
    if (!m.isContinuous()) {
        throw std::invalid_argument("Only continuous Mats supported.");
    }

    return py::array(determine_np_dtype(m.depth())
        , determine_shape(m)
        , m.data
        , make_capsule(m));
}
Run Code Online (Sandbox Code Playgroud)

让我们假设,我们有一个 Python 函数,如

def foo(arr):
    print(arr.shape)
Run Code Online (Sandbox Code Playgroud)

在 pybind 对象中捕获fun。然后要使用 aMat作为源从 C++ 调用此函数,我们将执行以下操作:

cv::Mat img; // Initialize this somehow

auto result = fun(mat_to_nparray(img));
Run Code Online (Sandbox Code Playgroud)

示例程序

#include <pybind11/pybind11.h>
#include <pybind11/embed.h>
#include <pybind11/numpy.h>
#include <pybind11/stl.h>

#include <opencv2/opencv.hpp>

#include <iostream>

namespace py = pybind11;

// The 4 functions from above go here...

int main()
{
    // Start the interpreter and keep it alive
    py::scoped_interpreter guard{};

    try {
        auto locals = py::dict{};

        py::exec(R"(
            import numpy as np

            def test_cpp_to_py(arr):
                return (arr[0,0,0], 2.0, 30)
        )");

        auto test_cpp_to_py = py::globals()["test_cpp_to_py"];


        for (int i = 0; i < 10; i++) {
            int64 t0 = cv::getTickCount();

            cv::Mat img(cv::Mat::zeros(1024, 1024, CV_8UC3) + cv::Scalar(1, 1, 1));

            int64 t1 = cv::getTickCount();

            auto result = test_cpp_to_py(mat_to_nparray(img));

            int64 t2 = cv::getTickCount();

            double delta0 = (t1 - t0) / cv::getTickFrequency() * 1000;
            double delta1 = (t2 - t1) / cv::getTickFrequency() * 1000;

            std::cout << "* " << delta0 << " ms | " << delta1 << " ms" << std::endl;
        }        
    } catch (py::error_already_set& e) {
        std::cerr << e.what() << "\n";
    }

    return 0;
}
Run Code Online (Sandbox Code Playgroud)

控制台输出

* 4.56413 ms | 0.225657 ms
* 3.95923 ms | 0.0736127 ms
* 3.80335 ms | 0.0438603 ms
* 3.99262 ms | 0.0577587 ms
* 3.82262 ms | 0.0572 ms
* 3.72373 ms | 0.0394603 ms
* 3.74014 ms | 0.0405079 ms
* 3.80621 ms | 0.054546 ms
* 3.72177 ms | 0.0386222 ms
* 3.70683 ms | 0.0373651 ms
Run Code Online (Sandbox Code Playgroud)