• opencv中两个LSD直线检测算法的区别与应用

    opencv中两个LSD直线检测算法的区别与应用

    同样是Line Segment Detector(lsd)算法，opencv中提供了两种实现，并且位于不同的模块。下面分别介绍它们的使用方法：

    1. LineSegmentDetector

    由于源码许可证问题 OpenCV 3.4.6-3.4.15、4.1.0-4.5.3中无法使用这个方法

    使用该类检测直线并显示的代码如下：

    import cv2

 if __name__ == '__main__':
    img = cv2.imread("test.jpg")
    gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    gray_img = cv2.GaussianBlur(gray_img, (3,3), 2.0)
    #LineSegmentDetector
    lsd = cv2.createLineSegmentDetector(refine=cv2.LSD_REFINE_NONE, scale=0.8, ang_th=35)
    lines_detected, width, prec, nfa = lsd.detect(gray_img)
    #lsd.drawSegments(img,lines_detected)
    #绘制检测结果
    for dline in lines_detected:
        x0 = int(round(dline[0][0]))
        y0 = int(round(dline[0][1]))
        x1 = int(round(dline[0][2]))
        y1 = int(round(dline[0][3]))
        cv2.line(mask, (x0, y0), (x1,y1), 255, 1, cv2.LINE_AA)
    cv2.imshow("Detected lines", img)
    cv2.waitKey(0)
    cv.destroyAllWindows()

    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18
    • 19
    • 20

    c++示例代码如下(lsd_lines.cpp)：

    #include "opencv2/imgproc.hpp"
#include "opencv2/imgcodecs.hpp"
#include "opencv2/highgui.hpp"
#include 
using namespace std;
using namespace cv;
int main(int argc, char** argv)
{
    cv::CommandLineParser parser(argc, argv,
                                 "{input   i|../data/building.jpg|input image}"
                                 "{refine  r|false|if true use LSD_REFINE_STD method, if false use LSD_REFINE_NONE method}"
                                 "{canny   c|false|use Canny edge detector}"
                                 "{overlay o|false|show result on input image}"
                                 "{help    h|false|show help message}");
    if (parser.get<bool>("help"))
    {
        parser.printMessage();
        return 0;
    }
    parser.printMessage();
    String filename = parser.get<String>("input");
    bool useRefine = parser.get<bool>("refine");
    bool useCanny = parser.get<bool>("canny");
    bool overlay = parser.get<bool>("overlay");
    Mat image = imread(filename, IMREAD_GRAYSCALE);
    if( image.empty() )
    {
        cout << "Unable to load " << filename;
        return 1;
    }
    imshow("Source Image", image);
    if (useCanny)
    {
        Canny(image, image, 50, 200, 3); // Apply Canny edge detector
    }
    // Create and LSD detector with standard or no refinement.
    Ptr<LineSegmentDetector> ls = useRefine ? createLineSegmentDetector(LSD_REFINE_STD) : createLineSegmentDetector(LSD_REFINE_NONE);
    double start = double(getTickCount());
    vector<Vec4f> lines_std;
    // Detect the lines
    ls->detect(image, lines_std);
    double duration_ms = (double(getTickCount()) - start) * 1000 / getTickFrequency();
    std::cout << "It took " << duration_ms << " ms." << std::endl;
    // Show found lines
    if (!overlay || useCanny)
    {
        image = Scalar(0, 0, 0);
    }
    ls->drawSegments(image, lines_std);
    String window_name = useRefine ? "Result - standard refinement" : "Result - no refinement";
    window_name += useCanny ? " - Canny edge detector used" : "";
    imshow(window_name, image);
    waitKey();
    return 0;
}

    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18
    • 19
    • 20
    • 21
    • 22
    • 23
    • 24
    • 25
    • 26
    • 27
    • 28
    • 29
    • 30
    • 31
    • 32
    • 33
    • 34
    • 35
    • 36
    • 37
    • 38
    • 39
    • 40
    • 41
    • 42
    • 43
    • 44
    • 45
    • 46
    • 47
    • 48
    • 49
    • 50
    • 51
    • 52
    • 53
    • 54
    • 55

    2. cv::line_descriptor::LSDDetector

    这个类在opencv_contrib中实现。源码目录结构示例如下：在这里插入图片描述
    如果c++编译opencv时未包含contrib模块；或在python中用pip安装了opencv-python而不是opencv-contrib-python，均无法使用该LSD直线检测方法。
    python用户只管卸载opencv-python，安装opencv-contrib-python，前者是后者的子集。

    pip uninstall opencv-python
pip install opencv-contrib-python

    • 1
    • 2

    使用该类检测直线并显示的python代码如下：

    import cv2

 if __name__ == '__main__':
    img = cv2.imread("test.jpg")
    gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    gray_img = cv2.GaussianBlur(gray_img, (3,3), 2.0)
    #LSDDetector
    lsd = cv2.line_descriptor_LSDDetector.createLSDDetector()
    lines = lsd.detect(gray, 2, 1)
    for kl in lines:#绘制检测结果
        if kl.octave == 0:
            # cv.line only accepts integer coordinate
            pt1 = (int(kl.startPointX), int(kl.startPointY))
            pt2 = (int(kl.endPointX), int(kl.endPointY))
            cv.line(img, pt1, pt2, [255, 0, 0], 2)
    cv.imshow('Detected lines', img)
    cv.waitKey(0)
    cv.destroyAllWindows()

    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18

    c++示例代码如下(lines_extraction.cpp)：

    #include 
#include 
#include 
#include 
#include 
#include 
#include 

using namespace cv;
using namespace cv::line_descriptor;
using namespace std;

static const char* keys =
{ "{@image_path | | Image path }" };

static void help()
{
  cout << "\nThis example shows the functionalities of lines extraction " << "furnished by BinaryDescriptor class\n"
       << "Please, run this sample using a command in the form\n" << "./example_line_descriptor_lines_extraction " << endl;
}

int main( int argc, char** argv )
{
  /* get parameters from comand line */
  CommandLineParser parser( argc, argv, keys );
  String image_path = parser.get<String>( 0 );

  if( image_path.empty() )
  {
    help();
    return -1;
  }

  /* load image */
  cv::Mat imageMat = imread( image_path, 1 );
  if( imageMat.data == NULL )
  {
    std::cout << "Error, image could not be loaded. Please, check its path" << std::endl;
    return -1;
  }

  /* create a random binary mask */
  cv::Mat mask = Mat::ones( imageMat.size(), CV_8UC1 );

  /* create a pointer to a BinaryDescriptor object with deafult parameters */
  Ptr<LSDDetector> bd = LSDDetector::createLSDDetector();

  /* create a structure to store extracted lines */
  vector<KeyLine> lines;

  /* extract lines */
  cv::Mat output = imageMat.clone();
  bd->detect( imageMat, lines, 2, 1, mask );

  /* draw lines extracted from octave 0 */
  if( output.channels() == 1 )
    cvtColor( output, output, COLOR_GRAY2BGR );
  for ( size_t i = 0; i < lines.size(); i++ )
  {
    KeyLine kl = lines[i];
    if( kl.octave == 0)
    {
      /* get a random color */
      int R = ( rand() % (int) ( 255 + 1 ) );
      int G = ( rand() % (int) ( 255 + 1 ) );
      int B = ( rand() % (int) ( 255 + 1 ) );

      /* get extremes of line */
      Point pt1 = Point2f( kl.startPointX, kl.startPointY );
      Point pt2 = Point2f( kl.endPointX, kl.endPointY );

      /* draw line */
      line( output, pt1, pt2, Scalar( B, G, R ), 3 );
    }

  }

  /* show lines on image */
  imshow( "LSD lines", output );
  waitKey();
}


    • 1
    • 2
    • 3
    • 4
    • 5
    • 6
    • 7
    • 8
    • 9
    • 10
    • 11
    • 12
    • 13
    • 14
    • 15
    • 16
    • 17
    • 18
    • 19
    • 20
    • 21
    • 22
    • 23
    • 24
    • 25
    • 26
    • 27
    • 28
    • 29
    • 30
    • 31
    • 32
    • 33
    • 34
    • 35
    • 36
    • 37
    • 38
    • 39
    • 40
    • 41
    • 42
    • 43
    • 44
    • 45
    • 46
    • 47
    • 48
    • 49
    • 50
    • 51
    • 52
    • 53
    • 54
    • 55
    • 56
    • 57
    • 58
    • 59
    • 60
    • 61
    • 62
    • 63
    • 64
    • 65
    • 66
    • 67
    • 68
    • 69
    • 70
    • 71
    • 72
    • 73
    • 74
    • 75
    • 76
    • 77
    • 78
    • 79
    • 80
    • 81
    • 82

    3. 区别与应用

    这两种LSD实现的处理结果基本一致，但是获取检测到的直线的端点坐标的方式有所不同：
    前者的直线检测结果是numpy array类型，每个直线由4个浮点数表示(x1, y1, x2, y2)；后者的检测结果的直线类型，具有起点和端点成员变量，分别是(startPointX,startPointY,endPointX,endPointY)，并且还具有直线的方向（角度）和长度成员变量。
    具体从上面两个代码片段的可视化部分展示的清清楚楚。

  • 相关阅读:
    VLAN笔记
    《大模型时代-ChatGPT开启通用人工智能浪潮》精华摘抄
    渗透测试-Python安全工具编程进阶
    Unity微信小游戏登录授权获取用户信息
    Java基础知识总结（42）
    php的短信验证的流程,如何实现前端js加后端php
    el-table 表格表头、单元格、滚动条样式修改
    C++ -- IO流
    java计算机毕业设计固定资产管理系统源码+mysql数据库+系统+LW文档+部署
    【TypeScript】语法详解 - 类型操作的补充
  • 原文地址：https://blog.csdn.net/qq_42679415/article/details/136342328