一、簡介

霍夫變換(Hough Transform)是影象處理中的一種特徵提取技術，它通過一種投票演算法檢測具有特定形狀的物體。該過程在一個引數空間中通過計算累計結果的區域性最大值得到一個符合該特定形狀的集合作為霍夫變換結果。霍夫變換於1962年由Paul Hough 首次提出[53]，後於1972年由Richard Duda和Peter Hart推廣使用[54]，經典霍夫變換用來檢測影象中的直線，後來霍夫變換擴充套件到任意形狀物體的識別，多為圓和橢圓。

霍夫變換運用兩個座標空間之間的變換將在一個空間中具有相同形狀的曲線或直線對映到另一個座標空間的一個點上形成峰值，從而把檢測任意形狀的問題轉化為統計峰值問題，上一節中已經介紹了車道的直線特徵，本節中介紹hough變換檢測直線的原理和檢測結果。

我們知道，一條直線在直角座標系下可以用y=kx+b表示, 霍夫變換的主要思想是將該方程的引數和變數交換，即用x,y作為已知量k,b作為變數座標，所以直角座標系下的直線y=kx+b在引數空間表示為點(k,b)，而一個點(x1,y1)在直角座標系下表示為一條直線y1=x1·k+b，其中(k,b)是該直線上的任意點。為了計算方便，我們將引數空間的座標表示為極座標下的γ和θ。因為同一條直線上的點對應的(γ,θ)是相同的，因此可以先將圖片進行邊緣檢測，然後對影象上每一個非零畫素點，在引數座標下變換為一條直線，那麼在直角座標下屬於同一條直線的點便在引數空間形成多條直線並內交於一點。因此可用該原理進行直線檢測。

如圖 所示，對於原圖內任一點(x,y)都可以在引數空間形成一條直線，以圖中一條直線為例有引數(γ,θ)=(69.641,30°)，所有屬於同一條直線上的點會在引數空間交於一點，該點即為對應直線的引數。由該圖中所有直線所得到的(γ,θ)在引數空間中得到一系列對應曲線見圖 霍夫統計變換結果。

二、原始碼

function varargout = HoughObject(varargin)
% HOUGHOBJECT M-file for HoughObject.fig
%      HOUGHOBJECT, by itself, creates a new HOUGHOBJECT or raises the existing
%      singleton*.
%
%      H = HOUGHOBJECT returns the handle to a new HOUGHOBJECT or the handle to
%      the existing singleton*.
%
%      HOUGHOBJECT('CALLBACK',hObject,eventData,handles,...) calls the local
%      function named CALLBACK in HOUGHOBJECT.M with the given input arguments.
%
%      HOUGHOBJECT('Property','Value',...) creates a new HOUGHOBJECT or raises the
%      existing singleton*.  Starting from the left, property value pairs are
%      applied to the GUI before HoughObject_OpeningFunction gets called.  An
%      unrecognized property name or invalid value makes property application
%      stop.  All inputs are passed to HoughObject_OpeningFcn via varargin.
%
%      *See GUI Options on GUIDE's Tools menu.  Choose "GUI allows only one
%      instance to run (singleton)".
%
% See also: GUIDE, GUIDATA, GUIHANDLES

% Edit the above text to modify the response to help HoughObject

% Last Modified by GUIDE v2.5 07-Dec-2005 20:15:07

% Begin initialization code - DO NOT EDIT
gui_Singleton = 1;
gui_State = struct('gui_Name',       mfilename, ...
                   'gui_Singleton',  gui_Singleton, ...
                   'gui_OpeningFcn', @HoughObject_OpeningFcn, ...
                   'gui_OutputFcn',  @HoughObject_OutputFcn, ...
                   'gui_LayoutFcn',  [] , ...
                   'gui_Callback',   []);
if nargin && ischar(varargin{1})
    gui_State.gui_Callback = str2func(varargin{1});
end

if nargout
    [varargout{1:nargout}] = gui_mainfcn(gui_State, varargin{:});
else
    gui_mainfcn(gui_State, varargin{:});
end
% End initialization code - DO NOT EDIT


% --- Executes just before HoughObject is made visible.
function HoughObject_OpeningFcn(hObject, eventdata, handles, varargin)
% This function has no output args, see OutputFcn.
% hObject    handle to figure
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)
% varargin   command line arguments to HoughObject (see VARARGIN)

% Choose default command line output for HoughObject
handles.output = hObject;

% Update handles structure
guidata(hObject, handles);

% UIWAIT makes HoughObject wait for user response (see UIRESUME)
% uiwait(handles.figure1);


% --- Outputs from this function are returned to the command line.
function varargout = HoughObject_OutputFcn(hObject, eventdata, handles) 
% varargout  cell array for returning output args (see VARARGOUT);
% hObject    handle to figure
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)

% Get default command line output from handles structure
varargout{1} = handles.output;


% --- Executes on button press in pbGenerate.
function pbGenerate_Callback(hObject, eventdata, handles)
% hObject    handle to pbGenerate (see GCBO)
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)
load ObjectTemplate
S1 = zeros(180,180);
Str = {'Sq','Sc','St'};

a = 1; b = 3;
x = round(a + (b-a) * rand(1));
S2 = eval(Str{x});

a = 1; b = 90;
x = round(a + (b-a) * rand(1));
S3 = imrotate(S2,x);

S = S1;
sz = size(S3);

a = 1; b = min(sz);
x = round(a + (b-a) * rand(1));


S(x:sz(1)+x-1,x:sz(2)+x-1)=S3;
S = im2bw(S);
axes(handles.axes1);

imshow(S);
title('Original Image');
handles.S = S;
guidata(hObject, handles);

 

三、執行結果

四、備註

版本：2014a
完整程式碼或代寫加1564658423