申明:本文非筆者原創,原文轉載自:http://blog.csdn.net/zouxy09/article/details/9622285
因為監控發展的需求,目前前景檢測的研究還是很多的,也出現了很多新的方法和思路。個人了解的大概概括為以下一些:
幀差、背景減除(GMM、CodeBook、 SOBS、 SACON、 VIBE、 W4、多幀平均……)、光流(稀疏光流、稠密光流)、運動競争(Motion Competition)、運動模版(運動曆史圖像)、時間熵……等等。如果加上他們的改進版,那就是很大的一個家族了。
對于上一些方法的一點簡單的對比分析可以參考下:
http://www.cnblogs.com/ronny/archive/2012/04/12/2444053.html
至于哪個最好,看使用環境吧,各有千秋,有一些适用的情況更多,有一些在某些情況下表現更好。這些都需要針對自己的使用情況作測試确定的。呵呵。
推薦一個牛逼的庫:http://code.google.com/p/bgslibrary/裡面包含了各種背景減除的方法,可以讓自己少做很多力氣活。
還有王先榮部落格上存在不少的分析:
http://www.cnblogs.com/xrwang/archive/2010/02/21/ForegroundDetection.html
下面的部落格上轉載王先榮的上面幾篇,然後加上自己分析了兩篇:
http://blog.csdn.net/stellar0
本文主要關注其中的一種背景減除方法:ViBe。stellar0的部落格上對ViBe進行了分析,我這裡就不再啰嗦了,具體的理論可以參考:
http://www2.ulg.ac.be/telecom/research/vibe/
http://blog.csdn.net/stellar0/article/details/8777283
http://blog.csdn.net/yongshengsilingsa/article/details/6659859
http://www2.ulg.ac.be/telecom/research/vibe/download.html
http://www.cvchina.info/2011/12/25/vibe/
《ViBe: A universal background subtraction algorithm for video sequences》
《ViBe: a powerful technique for background detection and subtraction in video sequences》
ViBe是一種像素級視訊背景模組化或前景檢測的算法,效果優于所熟知的幾種算法,對硬體記憶體占用也少,很簡單。我之前根據stellar0的代碼(在這裡,非常感謝stellar0)改寫成一個Mat格式的代碼了,現在擺上來和大家交流,具體如下:(在VS2010+OpenCV2.4.2中測試通過)
ViBe.h
[cpp] view plain copy
- #pragma once
- #include <iostream>
- #include "opencv2/opencv.hpp"
- using namespace cv;
- using namespace std;
- #define NUM_SAMPLES 20 //每個像素點的樣本個數
- #define MIN_MATCHES 2 //#min指數
- #define RADIUS 20 //Sqthere半徑
- #define SUBSAMPLE_FACTOR 16 //子采樣機率
- class ViBe_BGS
- {
- public:
- ViBe_BGS(void);
- ~ViBe_BGS(void);
- void init(const Mat _image); //初始化
- void processFirstFrame(const Mat _image);
- void testAndUpdate(const Mat _image); //更新
- Mat getMask(void){return m_mask;};
- private:
- Mat m_samples[NUM_SAMPLES];
- Mat m_foregroundMatchCount;
- Mat m_mask;
- };
ViBe.cpp
[cpp] view plain copy
- #include <opencv2/opencv.hpp>
- #include <iostream>
- #include "ViBe.h"
- using namespace std;
- using namespace cv;
- int c_xoff[9] = {-1, 0, 1, -1, 1, -1, 0, 1, 0}; //x的鄰居點
- int c_yoff[9] = {-1, 0, 1, -1, 1, -1, 0, 1, 0}; //y的鄰居點
- ViBe_BGS::ViBe_BGS(void)
- {
- }
- ViBe_BGS::~ViBe_BGS(void)
- {
- }
- void ViBe_BGS::init(const Mat _image)
- {
- for(int i = 0; i < NUM_SAMPLES; i++)
- {
- m_samples[i] = Mat::zeros(_image.size(), CV_8UC1);
- }
- m_mask = Mat::zeros(_image.size(),CV_8UC1);
- m_foregroundMatchCount = Mat::zeros(_image.size(),CV_8UC1);
- }
- void ViBe_BGS::processFirstFrame(const Mat _image)
- {
- RNG rng;
- int row, col;
- for(int i = 0; i < _image.rows; i++)
- {
- for(int j = 0; j < _image.cols; j++)
- {
- for(int k = 0 ; k < NUM_SAMPLES; k++)
- {
- // Random pick up NUM_SAMPLES pixel in neighbourhood to construct the model
- int random = rng.uniform(0, 9);
- row = i + c_yoff[random];
- if (row < 0)
- row = 0;
- if (row >= _image.rows)
- row = _image.rows - 1;
- col = j + c_xoff[random];
- if (col < 0)
- col = 0;
- if (col >= _image.cols)
- col = _image.cols - 1;
- m_samples[k].at<uchar>(i, j) = _image.at<uchar>(row, col);
- }
- }
- }
- }
- void ViBe_BGS::testAndUpdate(const Mat _image)
- {
- RNG rng;
- for(int i = 0; i < _image.rows; i++)
- {
- for(int j = 0; j < _image.cols; j++)
- {
- int matches(0), count(0);
- float dist;
- while(matches < MIN_MATCHES && count < NUM_SAMPLES)
- {
- dist = abs(m_samples[count].at<uchar>(i, j) - _image.at<uchar>(i, j));
- if (dist < RADIUS)
- matches++;
- count++;
- }
- if (matches >= MIN_MATCHES)
- {
- // It is a background pixel
- m_foregroundMatchCount.at<uchar>(i, j) = 0;
- // Set background pixel to 0
- m_mask.at<uchar>(i, j) = 0;
- // 如果一個像素是背景點,那麼它有 1 / defaultSubsamplingFactor 的機率去更新自己的模型樣本值
- int random = rng.uniform(0, SUBSAMPLE_FACTOR);
- if (random == 0)
- {
- random = rng.uniform(0, NUM_SAMPLES);
- m_samples[random].at<uchar>(i, j) = _image.at<uchar>(i, j);
- }
- // 同時也有 1 / defaultSubsamplingFactor 的機率去更新它的鄰居點的模型樣本值
- random = rng.uniform(0, SUBSAMPLE_FACTOR);
- if (random == 0)
- {
- int row, col;
- random = rng.uniform(0, 9);
- row = i + c_yoff[random];
- if (row < 0)
- row = 0;
- if (row >= _image.rows)
- row = _image.rows - 1;
- random = rng.uniform(0, 9);
- col = j + c_xoff[random];
- if (col < 0)
- col = 0;
- if (col >= _image.cols)
- col = _image.cols - 1;
- random = rng.uniform(0, NUM_SAMPLES);
- m_samples[random].at<uchar>(row, col) = _image.at<uchar>(i, j);
- }
- }
- else
- {
- // It is a foreground pixel
- m_foregroundMatchCount.at<uchar>(i, j)++;
- // Set background pixel to 255
- m_mask.at<uchar>(i, j) = 255;
- //如果某個像素點連續N次被檢測為前景,則認為一塊靜止區域被誤判為運動,将其更新為背景點
- if (m_foregroundMatchCount.at<uchar>(i, j) > 50)
- {
- int random = rng.uniform(0, NUM_SAMPLES);
- if (random == 0)
- {
- random = rng.uniform(0, NUM_SAMPLES);
- m_samples[random].at<uchar>(i, j) = _image.at<uchar>(i, j);
- }
- }
- }
- }
- }
- }
Main.cpp
[cpp] view plain copy
- // This is based on
- // "VIBE: A POWERFUL RANDOM TECHNIQUE TO ESTIMATE THE BACKGROUND IN VIDEO SEQUENCES"
- // by Olivier Barnich and Marc Van Droogenbroeck
- // Author : zouxy
- // Date : 2013-4-13
- // HomePage : http://blog.csdn.net/zouxy09
- // Email : [email protected]
- #include "opencv2/opencv.hpp"
- #include "ViBe.h"
- #include <iostream>
- #include <cstdio>
- using namespace cv;
- using namespace std;
- int main(int argc, char* argv[])
- {
- Mat frame, gray, mask;
- VideoCapture capture;
- capture.open("video.avi");
- if (!capture.isOpened())
- {
- cout<<"No camera or video input!\n"<<endl;
- return -1;
- }
- ViBe_BGS Vibe_Bgs;
- int count = 0;
- while (1)
- {
- count++;
- capture >> frame;
- if (frame.empty())
- break;
- cvtColor(frame, gray, CV_RGB2GRAY);
- if (count == 1)
- {
- Vibe_Bgs.init(gray);
- Vibe_Bgs.processFirstFrame(gray);
- cout<<" Training GMM complete!"<<endl;
- }
- else
- {
- Vibe_Bgs.testAndUpdate(gray);
- mask = Vibe_Bgs.getMask();
- morphologyEx(mask, mask, MORPH_OPEN, Mat());
- imshow("mask", mask);
- }
- imshow("input", frame);
- if ( cvWaitKey(10) == 'q' )
- break;
- }
- return 0;
- }