opencv图像拼接

本文代码加图片下载地址：
https://download.csdn.net/download/yang332233/12519478
图片如下：

先上拼接后的效果图：

本代码在opencv2.4.9编译通过，在opencv3.2会有报错，估计哪里函数改了。
OpenCV Error: The function/feature is not implemented (OpenCV was built without SURF support) in SurfFeaturesFinder, file /build/opencv-ys8xiq/opencv-2.4.9.1+dfsg/modules/stitching/src/matchers.cpp, line 319
qt中的pro文件如下：
.pro

QT -= gui

CONFIG += c++11 console
CONFIG -= app_bundle

# The following define makes your compiler emit warnings if you use
# any feature of Qt which as been marked deprecated (the exact warnings
# depend on your compiler). Please consult the documentation of the
# deprecated API in order to know how to port your code away from it.
DEFINES += QT_DEPRECATED_WARNINGS

# You can also make your code fail to compile if you use deprecated APIs.
# In order to do so, uncomment the following line.
# You can also select to disable deprecated APIs only up to a certain version of Qt.
#DEFINES += QT_DISABLE_DEPRECATED_BEFORE=0x060000    # disables all the APIs deprecated before Qt 6.0.0

SOURCES += main.cpp 

OPENCV_ROOT_PATH = /home/yhl/software/opencv2.4.9
#OPENCV_ROOT_PATH = /home/yhl/software

INCLUDEPATH += $${OPENCV_ROOT_PATH}/include 
               $${OPENCV_ROOT_PATH}/include/opencv 
               $${OPENCV_ROOT_PATH}/include/opencv2

LIBS += -L$${OPENCV_ROOT_PATH}/lib


LIBS += -lopencv_core 
        -lopencv_highgui 
        -lopencv_imgproc 
#        -lopencv_imgcodecs 
#        -lopencv_videoio 
        -lopencv_stitching 
        -lopencv_calib3d 
        -lopencv_stitching
        -lopencv_calib3d

//#include"stdafx.h"
#include <iostream>
#include <fstream>
#include <string>
#include "opencv2/opencv_modules.hpp"
#include "opencv2/highgui/highgui.hpp"
#include "opencv2/stitching/detail/autocalib.hpp"
#include "opencv2/stitching/detail/blenders.hpp"
#include "opencv2/stitching/detail/camera.hpp"
#include "opencv2/stitching/detail/exposure_compensate.hpp"
#include "opencv2/stitching/detail/matchers.hpp"
#include "opencv2/stitching/detail/motion_estimators.hpp"
#include "opencv2/stitching/detail/seam_finders.hpp"
#include "opencv2/stitching/detail/util.hpp"
#include "opencv2/stitching/detail/warpers.hpp"
#include "opencv2/stitching/warpers.hpp"

using namespace std;
using namespace cv;
using namespace cv::detail;

// Default command line args
vector<string> img_names;
bool preview = false;
bool try_gpu = false;
double work_megapix = 0.3;//图像的尺寸大小
double seam_megapix = 0.1;//拼接缝像素的大小
double compose_megapix = -1;//拼接分辨率
float conf_thresh = 1.f;//来自同一全景图置信度
string ba_cost_func = "ray";
string ba_refine_mask = "xxxxx";
bool do_wave_correct = true;
WaveCorrectKind wave_correct = detail::WAVE_CORRECT_HORIZ;//水平波形校检
bool save_graph = false;
std::string save_graph_to;
int expos_comp_type = ExposureCompensator::GAIN_BLOCKS;//光照补偿方法
float match_conf = 0.5f;//特征点检测置信等级，最近邻匹配距离和次近邻匹配距离比值
int blend_type = Blender::MULTI_BAND;//融合方法
float blend_strength = 5;
string result_name = "result.jpg";

int main(int argc, char* argv[])
{
#if ENABLE_LOG
    int64 app_start_time = getTickCount();
#endif

    /*int n2,n3;
    String str3 = "D://";
    cout<<"请输入第一个拼接视频序号"<<endl;
    cin>>n2;
    string str4,str5;
    stringstream stream3;
    stream3<<n2;
    stream3>>str4;
    str5=str3+str4+".mp4";
    const char* p1 = str5.data();//
    cout<<str5<<endl;
    int period =42,count=1; //帧间隔，每隔多少帧取其中一张截图

    CvCapture *capture1 = cvCaptureFromAVI(p1);
    int start_frame1=120,end_frame1=1900;
    cvSetCaptureProperty(capture1,CV_CAP_PROP_POS_FRAMES,start_frame1);
    char filename1[128];
    if (capture1 == NULL)
    return -1;

    IplImage *frame1;
    while(cvGrabFrame(capture1) && start_frame1 <= end_frame1)
    {
        for(int i=0;i<period;i++)
        {
            frame1=cvQueryFrame(capture1);
            if(!frame1)
            {
        printf("finish!
");
             break;

             }
        }
        sprintf(filename1,"D://4//img_%d.jpg",count++);
        cvSaveImage(filename1,frame1);}
    cout<<"读取第一个视频完毕"<<endl;
    cout<<"请输入第二个拼接视频序号"<<endl;
    int count1=1;
    cin>>n3;
    string str_4,str_5;
    stringstream stream3_;
    stream3_<<n3;
    stream3_>>str_4;
    str_5=str3+str_4+".mp4";
     //帧间隔，每隔多少帧取其中一张截图
    const char* p2 = str_5.data();//
    CvCapture *capture = cvCaptureFromAVI(p2);
    int start_frame=120,end_frame=1900;
    cvSetCaptureProperty(capture,CV_CAP_PROP_POS_FRAMES,start_frame);

    char filename[128];
    if (capture == NULL)
    return -1;

    IplImage *frame;
    while(cvGrabFrame(capture) && start_frame <= end_frame)
    {
        for(int i=0;i<period;i++)
        {
            frame=cvQueryFrame(capture);
            if(!frame)
            {
        printf("finish!
");
      break;
         return 0;
             }
        }
        sprintf(filename,"D://5//img_%d.jpg",count1++);
        cvSaveImage(filename,frame);}
    cout<<"读取第二个视频完毕"<<endl;*/

    int n;
    int n1;
    cout << "请输入拼接7度图片的数量" << endl;
   // cin >> n;
    cout << "请输入拼接5度图片的数量" << endl;
   // cin >> n1;
    n = 5;
    n1 = 5;
    String str = "D://4//img (";
    String str1 = ").jpg";
    String str2, str6;
    String str_= "D://5//img (";
    String str1_ = ").jpg";
    String str2_, str3_;
    for (int i =1; i <= n1; i++)
    {
        stringstream stream,stream1;
        int j=i+2;
        stream <<j;
        stream >> str6;
        stream1<<i;
        stream1>>str3_;
        str2 = str + str6 + str1;
        str2_ = str_ + str3_ + str1_;
        img_names.push_back(str2);
        img_names.push_back(str2_);
    }
    cv::setBreakOnError(true);

    img_names.clear();
    img_names.push_back("/media/d_2/everyday/0507/matlab/0506/TU/1.jpg");
    img_names.push_back("/media/d_2/everyday/0507/matlab/0506/TU/2.jpg");
    img_names.push_back("/media/d_2/everyday/0507/matlab/0506/TU/3.jpg");
    img_names.push_back("/media/d_2/everyday/0507/matlab/0506/TU/4.jpg");
    img_names.push_back("/media/d_2/everyday/0507/matlab/0506/TU/5.jpg");

    // Check if have enough images
    int num_images = static_cast<int>(img_names.size());
    if (num_images < 2)
    {
        cout<<"需要更多图片"<<endl;
        return -1;
    }
    //若按原图片大小进行拼接时，会在曝光补偿时造成内存溢出，所以在计算时缩小图像尺寸变为work_megapix*10000
    double work_scale = 1, seam_scale = 1, compose_scale = 1;
    bool is_work_scale_set = false, is_seam_scale_set = false, is_compose_scale_set = false;

    cout<<"寻找特征点..."<<endl;
#if 1//ENABLE_LOG
    int64 t = getTickCount();
#endif

    Ptr<FeaturesFinder> finder;
    finder = new SurfFeaturesFinder();//使用surf方法进行特征点检测

    vector<Mat> full_img(num_images);
    vector<Mat> img(num_images);
    vector<ImageFeatures> features(num_images);
    vector<Mat> images(num_images);
    vector<Size> full_img_sizes(num_images);
    double seam_work_aspect = 1;
    #pragma omp parallel for
    for (int i = 0; i < num_images; ++i)
    {
        full_img[i] = imread(img_names[i]);
        full_img_sizes[i] = full_img[i].size();
       //计算work_scale，将图像resize到面积在work_megapix*10^6以下
        work_scale = min(1.0, sqrt(work_megapix * 1e6 / full_img[i].size().area()));

        resize(full_img[i], img[i], Size(), work_scale, work_scale);
        //将图像resize到面积在work_megapix*10^6以下
        seam_scale = min(1.0, sqrt(seam_megapix * 1e6 / full_img[i].size().area()));
        seam_work_aspect = seam_scale / work_scale;
        // 计算图像特征点，以及计算特征点描述子，并将img_idx设置为i
        (*finder)(img[i], features[i]);
        features[i].img_idx = i;
        cout << "Features in image #" << i + 1 << ": " << features[i].keypoints.size() << endl;
        //将源图像resize到seam_megapix*10^6，并存入image[]中
        resize(full_img[i], img[i], Size(), seam_scale, seam_scale);
        images[i] = img[i].clone();

        resize(full_img[i], img[i], Size(), seam_scale, seam_scale);
        images[i] = img[i].clone();
    }

    finder->collectGarbage();
    cout<<"寻找特征点所用时间: " << ((getTickCount() - t) / getTickFrequency()) << " sec"<<endl;
    //对图像进行两两匹配
    cout << "两两匹配" << endl;
    t = getTickCount();
    //使用最近邻和次近邻匹配，对任意两幅图进行匹配
    vector<MatchesInfo> pairwise_matches;
    BestOf2NearestMatcher matcher(try_gpu, match_conf);

    matcher(features, pairwise_matches);//对pairwise_matches中大于0的进行匹配
    matcher.collectGarbage();
    cout<<"两两匹配所用时间: " << ((getTickCount() - t) / getTickFrequency()) << " sec"<<endl;
    for(int i=0;i<num_images * num_images;++i)
    {
        if(pairwise_matches[i].confidence==0)
            pairwise_matches[i].confidence=-1;

    }

    //将置信度高于设置门限的留下来，其他的删除
    vector<int> indices = leaveBiggestComponent(features, pairwise_matches, conf_thresh);//留下来的序号
    vector<Mat> img_subset;
    vector<string> img_names_subset;
    vector<Size> full_img_sizes_subset;
    for (size_t i = 0; i < indices.size(); ++i)
    {
        img_names_subset.push_back(img_names[indices[i]]);
        img_subset.push_back(images[indices[i]]);
        full_img_sizes_subset.push_back(full_img_sizes[indices[i]]);
    }

    images = img_subset;
    img_names = img_names_subset;
    full_img_sizes = full_img_sizes_subset;

    num_images = static_cast<int>(img_names.size());
    if (num_images < 2)
    {
        cout<<"需要更多图片"<<endl;
        return -1;
    }

    HomographyBasedEstimator estimator;//基于单应性矩阵H的估计量
    vector<CameraParams> cameras;//相机参数
    estimator(features, pairwise_matches, cameras);

    //计算出R矩阵
    for (size_t i = 0; i < cameras.size(); ++i)
    {
        Mat R;
        cameras[i].R.convertTo(R, CV_32F);
        cameras[i].R = R;
        cout << "Initial intrinsics #" << indices[i] + 1 << ":
" << cameras[i].K() << endl;
    }
    //捆绑调整方法精确求取变换参数
    Ptr<detail::BundleAdjusterBase> adjuster;
     adjuster = new detail::BundleAdjusterRay();//使用光束法平差对所有相机参数矫正

    adjuster->setConfThresh(conf_thresh);//设置阈值
    Mat_<uchar> refine_mask = Mat::zeros(3, 3, CV_8U);
    if (ba_refine_mask[0] == 'x') refine_mask(0, 0) = 1;
    if (ba_refine_mask[1] == 'x') refine_mask(0, 1) = 1;
    if (ba_refine_mask[2] == 'x') refine_mask(0, 2) = 1;
    if (ba_refine_mask[3] == 'x') refine_mask(1, 1) = 1;
    if (ba_refine_mask[4] == 'x') refine_mask(1, 2) = 1;
    adjuster->setRefinementMask(refine_mask);
    (*adjuster)(features, pairwise_matches, cameras);//cameras存储求精后的变换参数

    // Find median focal length

    vector<double> focals;//存入相机像素
    for (size_t i = 0; i < cameras.size(); ++i)
    {
        cout<<"Camera #" << indices[i] + 1 << ":
" << cameras[i].K()<<endl;
        focals.push_back(cameras[i].focal);
    }

    sort(focals.begin(), focals.end());
    float warped_image_scale;
    if (focals.size() % 2 == 1)
        warped_image_scale = static_cast<float>(focals[focals.size() / 2]);
    else
        warped_image_scale = static_cast<float>(focals[focals.size() / 2 - 1] + focals[focals.size() / 2]) * 0.5f;


        vector<Mat> rmats;
        for (size_t i = 0; i < cameras.size(); ++i)
            rmats.push_back(cameras[i].R);
        waveCorrect(rmats, wave_correct);//波形矫正
        for (size_t i = 0; i < cameras.size(); ++i)
            cameras[i].R = rmats[i];
        for (int i = 0; i < num_images; i++)
        cout << "R " << i <<":"<<cameras[i].R <<endl;

    cout << "弯曲图像 ... " << endl;
#if ENABLE_LOG
    t = getTickCount();
#endif

    vector<Point> corners(num_images);//统一坐标后的顶点
    vector<Mat> masks_warped(num_images);
    vector<Mat> images_warped(num_images);
    vector<Size> sizes(num_images);
    vector<Mat> masks(num_images);//融合掩码

    // 准备图像融合掩码
    for (int i = 0; i < num_images; ++i)
    {
        masks[i].create(images[i].size(), CV_8U);//masks为模，和图像一样大小，设置为白色，在上面进行融合
        masks[i].setTo(Scalar::all(255));
    }

    //弯曲图像和融合掩码

    Ptr<WarperCreator> warper_creator;
    warper_creator = new cv::SphericalWarper();//球面拼接

    vector<Mat> images_warped_f(num_images);
    Ptr<RotationWarper> warper = warper_creator->create(static_cast<float>(warped_image_scale * seam_work_aspect));//warped_image_scale焦距中值;
    #pragma omp parallel for
    for (int i = 0; i < num_images; ++i)
    {
        Mat_<float> K;
        cameras[i].K().convertTo(K, CV_32F);
        float swa = (float)seam_work_aspect;
        K(0,0) *= swa; K(0,2) *= swa;
        K(1,1) *= swa; K(1,2) *= swa;
        corners[i] = warper->warp(images[i], K, cameras[i].R, INTER_LINEAR, BORDER_REFLECT, images_warped[i]);
        cout << "k"<<i<<":" << K << endl;
        sizes[i] = images_warped[i].size();
        cout << "size" << i << ":" << sizes[i] << endl;
        cout << "corner" << i << ":" << corners[i] << endl;
        warper->warp(masks[i], K, cameras[i].R, INTER_NEAREST, BORDER_CONSTANT, masks_warped[i]);//膜进行变换
        images_warped[i].convertTo(images_warped_f[i], CV_32F);
    }


    cout<<"球面变换耗时" << ((getTickCount() - t) / getTickFrequency()) << " sec"<<endl;

    //查找图片之间的接缝
    Ptr<SeamFinder> seam_finder;
    seam_finder = new detail::GraphCutSeamFinder(GraphCutSeamFinderBase::COST_COLOR);
    seam_finder->find(images_warped_f, corners, masks_warped);

    //释放未使用内存
    images.clear();
    images_warped.clear();
    images_warped_f.clear();
    masks.clear();

    //////图像融合
    cout << "图像融合..." << endl;
#if ENABLE_LOG
    t = getTickCount();
#endif

    Mat img_warped, img_warped_s;
    Mat dilated_mask, seam_mask, mask, mask_warped;
    Ptr<Blender> blender;
    //double compose_seam_aspect = 1;
    double compose_work_aspect = 1;
    #pragma omp parallel for
    for (int img_idx = 0; img_idx < num_images; ++img_idx)
    {
        LOGLN("Compositing image #" << indices[img_idx] + 1);

        //重新计算

        if (!is_compose_scale_set)
        {
            if (compose_megapix > 0)
                compose_scale = min(1.0, sqrt(compose_megapix * 1e6 / full_img[img_idx].size().area()));
            is_compose_scale_set = true;


            compose_work_aspect = compose_scale / work_scale;


        // 更新弯曲图像比例
            warped_image_scale *= static_cast<float>(compose_work_aspect);
            warper = warper_creator->create(warped_image_scale);

            //更新 corners and sizes
            for (int i = 0; i < num_images; ++i)
            {
                // 更新相机属性
                cameras[i].focal *= compose_work_aspect;
                cameras[i].ppx *= compose_work_aspect;
                cameras[i].ppy *= compose_work_aspect;

                //更新 corner and size
                Size sz = full_img_sizes[i];
                if (std::abs(compose_scale - 1) > 1e-1)
                {
                    sz.width = cvRound(full_img_sizes[i].width * compose_scale);
                    sz.height = cvRound(full_img_sizes[i].height * compose_scale);
                }
                //corners和sizes
                Mat K;
                cameras[i].K().convertTo(K, CV_32F);
                Rect roi = warper->warpRoi(sz, K, cameras[i].R);
                corners[i] = roi.tl();
                sizes[i] = roi.size();
            }
        }
        if (abs(compose_scale - 1) > 1e-1)
            resize(full_img[img_idx], img[img_idx], Size(), compose_scale, compose_scale);
        else
            img = full_img;
        full_img[img_idx].release();
        Size img_size = img[img_idx].size();

        Mat K;
        cameras[img_idx].K().convertTo(K, CV_32F);

        // 扭曲当前图像
        warper->warp(img[img_idx], K, cameras[img_idx].R, INTER_LINEAR, BORDER_REFLECT, img_warped);

        // 扭曲当图像掩模
        mask.create(img_size, CV_8U);
        mask.setTo(Scalar::all(255));
        warper->warp(mask, K, cameras[img_idx].R, INTER_NEAREST, BORDER_CONSTANT, mask_warped);

        // 曝光补偿


        img_warped.convertTo(img_warped_s, CV_16S);
        img_warped.release();
        img[img_idx].release();
        mask.release();

        dilate(masks_warped[img_idx], dilated_mask, Mat());
        resize(dilated_mask, seam_mask, mask_warped.size());
        mask_warped = seam_mask & mask_warped;
        //初始化blender

        if (blender.empty())
        {
            blender = Blender::createDefault(blend_type, try_gpu);
            Size dst_sz = resultRoi(corners, sizes).size();
            float blend_width = sqrt(static_cast<float>(dst_sz.area())) * blend_strength / 100.f;

            if (blend_width < 1.f)
                blender = Blender::createDefault(Blender::NO, try_gpu);

            else
            {
                MultiBandBlender* mb = dynamic_cast<MultiBandBlender*>(static_cast<Blender*>(blender));
                mb->setNumBands(static_cast<int>(ceil(log(blend_width) / log(2.)) - 1.));

            }

        //根据corners顶点和图像的大小确定最终全景图的尺寸
            blender->prepare(corners, sizes);
        }

        // 融合当前图像
        blender->feed(img_warped_s, mask_warped, corners[img_idx]);
    }

    Mat result, result_mask;
    blender->blend(result, result_mask);

    namedWindow("result",WINDOW_NORMAL);
    imshow("result",result);
    waitKey(0);

    imwrite(result_name, result);
   // cout<<"程序运行时间为: " << ((getTickCount() - app_start_time) / getTickFrequency()) << " 秒"<<endl;
    return 0;
}