【图像处理】Hough变换直线检测

Matlab函数调用

% 入口图像为 BW，出口图像为f
clc,close
BW=imread('D:\picture\9dafa605d53eea243812bb29.jpg');
BW=rgb2gray(BW);
thresh=[0.01,0.17];
sigma=2;%定义高斯参数
f = edge(double(BW),'canny',thresh,sigma);
figure(1),imshow(f,[]);title('canny 边缘检测');
[H, theta, rho]= hough(f, 0.5);
%imshow(theta,rho,H,[],'notruesize'),
axis on,
axis normal
%xlabel('\theta'),ylabel('rho');
[r,c]=houghpeaks(H,5);
hold on
lines=houghlines(f,theta,rho,r,c);
figure,imshow(f,[]),title('Hough Transform Detect Result'),
hold on
for k=1:length(lines) 
    xy=[lines(k).point1;lines(k).point2];
plot(xy(:,1),xy(:,2),'LineWidth',4,'Color',[.6 .6 .6]);
end

##################################################

在上文中发现经过霍夫变换检测出的直线有可能因为车辆挡住路沿等原因断开，形成线段，这样就不好了，因为检测道路是要找直线焦点。

Thus it is necessary to combine 相同斜率的直线 and connect them.

本代码提供了matlab下求取经过霍夫变换的直线斜率，并将其联合，代码见下方，实验结果见文末。

% 入口图像为 BW，出口图像为f
%optimize from main_optimize, merely select 2 lines, one has positive
%slope,the other has negative slope
clear all,close all
BW=imread('road.jpg');
figure,imshow(BW);

BW=rgb2gray(BW);
%thresh=[0.01,0.17];
thresh=[0.01,0.10];
sigma=2;%定义高斯参数
f = edge(double(BW),'canny',thresh,sigma);
figure,subplot(121);
imshow(f,[]);
title('canny Edge Detect Result');

[H, theta, rho]= hough(f, 'Theta',20:0.1:75);%cos(theta)*x+sin(theta)*y=rho
%imshow(theta,rho,H,[],'notruesize'),axis on,axis normal
%xlabel('\theta'),ylabel('rho');

%求极值点
Peaks=houghpeaks(H,5);
%得到线段信息
lines=houghlines(f,theta,rho,Peaks);

% [r,c]=houghpeaks(H,10);
% hold on
% lines=houghlines(f,theta,rho,r,c);

subplot(122);
imshow(f,[]),title('Hough Transform Detect Result'),hold on
nlind=0;%new line index
st=1;
%%%%%%%%%求斜率%%%%%%%%%%%%
for k=1:length(lines)
    %xy=[lines(k).point1;lines(k).point2];
    xielv(k)=(lines(k).point2(1)-lines(k).point1(1))/(lines(k).point2(2)-lines(k).point1(2)+0.0001)
end

%%%%%%%%%将相同斜率的直线连起来%%%%%%%%%%%%
k=1;
while(k<=length(lines))
    if(k~=length(lines))
        k=k+1;
    end
    while(abs(xielv(k)-xielv(k-1))<0.0001)
        k=k+1;
        if(k>length(lines))
            break;
        end
    end

    if(abs(xielv(k-1))<0.05||abs(xielv(k-1))>=10)%eliminate horizontal and vertical lines,防治水平线和楼房
        st=k;
        if(k~=length(lines))
            continue;
        end
    end
    
    if(st==length(lines)&&k==st)
        if(abs(xielv(k))>0.05&&abs(xielv(k))<10)
            nlind=nlind+1;
            newlines(nlind)=lines(st);
            newlines(nlind).point2=lines(k).point2;
            newxy=[newlines(nlind).point1;newlines(nlind).point2];
            plot(newxy(:,2),newxy(:,1),'LineWidth',4,'Color',[.6 1.0 .8]);
        end
        break;
    end

    %end=k-1,start=st; draw line
    nlind=nlind+1;
    newlines(nlind)=lines(st);
    newlines(nlind).point2=lines(k-1).point2;
    newxy=[newlines(nlind).point1;newlines(nlind).point2];
    plot(newxy(:,2),newxy(:,1),'LineWidth',4,'Color',[.6 1.0 .8]);

    st=k;
end

fprintf('%d lines are detected in sum.\n',nlind);