Basler相机实时图像显示--Qt代码

Balser相机

前段时间，因为项目需要，师傅让我用Qt写巴斯勒相机实时画面的显示，网上找了好多资料，愚笨的我也没有弄明白该如何写，从相机的SDK中找到了用MFC写的例子，咱也没用过MFC呀，看不懂哎，最后师傅出马，写了一下，相机图像转换的位置师傅用他之前写的图像转换类，属于公司产品，我例子中就没用这个类，转而用opencv进行转换(网上找到的方法)，本文后面会附上源码。

本程序的环境：

Pylon 6.0.0 + Opencv 4.1.0 +Qt 5.12.3 + Vs 2017 Community

pylon 6.0.0 : Pylon 6.0 下载地址

vs下的头文件包含

下面只说明Balser的头文件包含，opencv 和qt的头文件包含这里忽略。

下载好之后，包含include

在其中填写如下路径：

XXX\Development\include
XXX\Development\Samples\C++\include

XXX\Development\Samples\C++\include路径也可以不包含，我是用来测试sample程序的，本源码就是仿照XXX/Development\Samples\C++\GUI_MFC这个Demo来写的。

下面来包含lib：

在其中填写如下路径：

XXX/Development\lib\x64 #我用的是X64版本的，X86请改为Win32

源码讲解：

main.c

#include "BaslerCamera_RealTimeShow.h"
#include <QtWidgets/QApplication>class PreWork
{
public:PreWork(){Pylon::PylonInitialize();//相机的初始化程序，把它理解为安装}~PreWork(){Pylon::PylonTerminate();//停止运行相机的执行函数，把它理解为卸载}
};int main(int argc, char *argv[])
{PreWork p;//主要说明这一点，巧用构造函数和析构函数，来对相机进行安装和卸载QApplication a(argc, argv);BaslerCamera_RealTimeShow w;w.show();return a.exec();
}

Basler.c

#include "BaslerCamera_RealTimeShow.h"
#include <QDebug>
#include <QPainter>
#include <QRect>BaslerCamera_RealTimeShow::BaslerCamera_RealTimeShow(QWidget *parent): QMainWindow(parent)
{ui.setupUi(this);m_camera.RegisterImageEventHandler(this, Pylon::RegistrationMode_ReplaceAll, Pylon::Ownership_ExternalOwnership);// Register this object as a configuration event handler, so we will be notified of camera state changes.// See Pylon::CConfigurationEventHandler for detailsm_camera.RegisterConfiguration(this, Pylon::RegistrationMode_ReplaceAll, Pylon::Ownership_ExternalOwnership);// Add the AutoPacketSizeConfiguration and let pylon delete it when not needed anymore.m_camera.RegisterConfiguration(new CAutoPacketSizeConfiguration(), Pylon::RegistrationMode_Append, Pylon::Cleanup_Delete);m_camera.Attach(Pylon::CTlFactory::GetInstance().CreateFirstDevice(), Pylon::Cleanup_Delete);m_camera.Open();// Camera may have been disconnected.if (!m_camera.IsOpen() || m_camera.IsGrabbing()){return;}// Since we may switch between single and continuous shot, we must configure the camera accordingly.// The predefined configurations are only executed once when the camera is opened.// To be able to use them in our use case, we just call them explicitly to apply the configuration.m_continousConfiguration.OnOpened(m_camera);// Start grabbing until StopGrabbing() is called.m_camera.StartGrabbing(Pylon::GrabStrategy_OneByOne, Pylon::GrabLoop_ProvidedByInstantCamera);ui.centralWidget->installEventFilter(this);//安装Qt的事件过滤器connect(this, SIGNAL(OneImageFinishSignal()), this, SLOT(OneImageFinishSlot()));}void BaslerCamera_RealTimeShow::OnImagesSkipped(Pylon::CInstantCamera& camera, size_t countOfSkippedImages)
{}
void BaslerCamera_RealTimeShow::OnImageGrabbed(Pylon::CInstantCamera& camera, const Pylon::CGrabResultPtr& grabResult)
{m_mutexLock.lock();m_ptrGrabResult = grabResult;//将捕获到的图像传递出去//qDebug() << __FUNCTION__;emit OneImageFinishSignal();m_mutexLock.unlock();}void BaslerCamera_RealTimeShow::OneImageFinishSlot()
{//qDebug() << __FUNCTION__;ui.centralWidget->update();
}
bool BaslerCamera_RealTimeShow::eventFilter(QObject *watched, QEvent *event)
{if (watched == ui.centralWidget && event->type() == QEvent::Paint){showImage();}return false;
}
void BaslerCamera_RealTimeShow::showImage()
{   m_mutexLock.lock();//qDebug() << "123" << endl;// 新建pylon ImageFormatConverter对象.CImageFormatConverter formatConverter;Mat openCvImage;QPainter painter(ui.centralWidget);//确定输出像素格式formatConverter.OutputPixelFormat = PixelType_BGR8packed;//将抓取的缓冲数据转化成pylon image.formatConverter.Convert(m_bitmapImage, m_ptrGrabResult);// 将 pylon image转成OpenCV image.openCvImage = cv::Mat(m_ptrGrabResult->GetHeight(), m_ptrGrabResult->GetWidth(), CV_8UC3, (uint8_t *)m_bitmapImage.GetBuffer());QImage img((const unsigned char *)(openCvImage.data), openCvImage.cols, openCvImage.rows, openCvImage.cols * 3, QImage::Format_RGB888);QRectF target;target.setLeft(0);target.setTop(0);target.setSize(this->size());QRectF source;source.setLeft(0);source.setTop(0);source.setSize(img.size());painter.drawImage(target,img, source);m_mutexLock.unlock();}// Pylon::CConfigurationEventHandler functions
void BaslerCamera_RealTimeShow::OnAttach(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnAttached(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnDetach(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnDetached(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnDestroy(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnDestroyed(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnOpen(Pylon::CInstantCamera& camera)
{Pylon::String_t strFriendlyName = camera.GetDeviceInfo().GetFriendlyName();qDebug() << __FUNCTION__ << " - " << strFriendlyName.c_str();
}void BaslerCamera_RealTimeShow::OnOpened(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnClose(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnClosed(Pylon::CInstantCamera& camera)
{Pylon::String_t strFriendlyName = camera.GetDeviceInfo().GetFriendlyName();qDebug() << __FUNCTION__ << " - " << strFriendlyName.c_str();
}void BaslerCamera_RealTimeShow::OnGrabStart(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnGrabStarted(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnGrabStop(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnGrabStopped(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;m_camera.DeregisterConfiguration(&m_continousConfiguration);
}void BaslerCamera_RealTimeShow::OnGrabError(Pylon::CInstantCamera& camera, const char* errorMessage)
{qDebug() << __FUNCTION__;
}void BaslerCamera_RealTimeShow::OnCameraDeviceRemoved(Pylon::CInstantCamera& camera)
{qDebug() << __FUNCTION__;
}BaslerCamera_RealTimeShow::~BaslerCamera_RealTimeShow()
{Perform cleanup.if (m_camera.IsPylonDeviceAttached()){try{// Close camera.// This will also stop the grab.m_camera.Close();// Free the camera.// This will also stop the grab and close the camera.m_camera.DestroyDevice();}catch (const Pylon::GenericException& e){qDebug() << e.what();}}
}

Balser.h

#pragma once#include <QtWidgets/QMainWindow>
#include "ui_BaslerCamera_RealTimeShow.h"
#include "AutoPacketSizeConfiguration.h"//加载PYLON API.
#include <pylon/PylonIncludes.h>
#include <pylon/gige/BaslerGigECamera.h>
#include <iostream>
#include <QMutex>// 加载OpenCV API
#include <opencv2/core/core.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/video/video.hpp>
#include <opencv2/opencv.hpp>//命名空间.
using namespace Pylon;
using namespace std;
using namespace cv;class BaslerCamera_RealTimeShow : public QMainWindow                                     , public Pylon::CImageEventHandler             // Allows you to get informed about received images and grab errors., public Pylon::CConfigurationEventHandler     // Allows you to get informed about device removal.{Q_OBJECTsignals:void OneImageFinishSignal();
private slots:void OneImageFinishSlot();public:BaslerCamera_RealTimeShow(QWidget *parent = Q_NULLPTR);~BaslerCamera_RealTimeShow();private:Ui::BaslerCamera_RealTimeShowClass ui;Pylon::CInstantCamera m_camera;// The grab result retrieved from the cameraPylon::CGrabResultPtr m_ptrGrabResult;// The grab result as a windows DIB to be displayed on the screenPylon::CPylonBitmapImage m_bitmapImage;Pylon::CAcquireContinuousConfiguration m_continousConfiguration;QMutex m_mutexLock;protected:void showImage();virtual bool eventFilter(QObject *watched, QEvent *event);// Pylon::CImageEventHandler functionsvirtual void OnImagesSkipped(Pylon::CInstantCamera& camera, size_t countOfSkippedImages);virtual void OnImageGrabbed(Pylon::CInstantCamera& camera, const Pylon::CGrabResultPtr& grabResult);// Pylon::CConfigurationEventHandler functionsvirtual void OnAttach(Pylon::CInstantCamera& camera);virtual void OnAttached(Pylon::CInstantCamera& camera);virtual void OnDetach(Pylon::CInstantCamera& camera);virtual void OnDetached(Pylon::CInstantCamera& camera);virtual void OnDestroy(Pylon::CInstantCamera& camera);virtual void OnDestroyed(Pylon::CInstantCamera& camera);virtual void OnOpen(Pylon::CInstantCamera& camera);virtual void OnOpened(Pylon::CInstantCamera& camera);virtual void OnClose(Pylon::CInstantCamera& camera);virtual void OnClosed(Pylon::CInstantCamera& camera);virtual void OnGrabStart(Pylon::CInstantCamera& camera);virtual void OnGrabStarted(Pylon::CInstantCamera& camera);virtual void OnGrabStop(Pylon::CInstantCamera& camera);virtual void OnGrabStopped(Pylon::CInstantCamera& camera);virtual void OnGrabError(Pylon::CInstantCamera& camera, const char* errorMessage);virtual void OnCameraDeviceRemoved(Pylon::CInstantCamera& camera);};

main.c内容很简单，执行安装和卸载的函数

Balser.c在这里着重说明一下：

先说一下Balser获取实时图像的原理吧，网上也有很多，这里按照我的理解写一下。

1、Basler获取实时图像是一张一张的拍照，拍照-显示-拍照-显示…，所以我们看到的就是连续的图像。

2、Basler是事件驱动的，当拍完一张照片之后，会触发事件，然后执行的Balser提供的OnImageGrabbed函数，这个原理可以理解为Qt的信号-槽

Balser.c的程序流程

捕获图像 —> 触发信号(事件)—> 执行OnImageGrabbed（）函数，将图像保存到我们定义的类中(这里注意下m_mutexLock.lock();的使用，因为巴斯勒相机处理图像是多线程的，所以在使用变量的时候，要用线程锁保护)—>发送我们自定义的信号-因为构造函数已经连接好了信号与槽-执行OneImageFinishSlot（）中的updata触发 event事件，判断该窗体时centralWidget，事件为QEvent::Paint，执行显示函数showImage（）。显示函数就是把图像转换一下格式，用Painter画一下图（也用到了线程锁）

最后的源码:

总结一下，本人实力有限，写的比较啰嗦，希望大家能看懂。

如果大家下载源码，附上CSDN和Gitee链接，如果大家CSDN积分多，就赞助点小弟，如果没有积分，请使用Gitee链接。

最后附上下载链接：

CSDN链接地址:CSDN

Gitee链接地址：Gitee