免规则采集器列表算法(算法代码主要在CMT这个类里面主要包含initialize和processFrame两个函数)

　　免规则采集器列表算法(算法代码主要在CMT这个类里面主要包含initialize和processFrame两个函数)

　　cmt 算法 cmt 算法

　　代码主要在CMT类中，主要包括initialize和processFrame函数的实现。它非常适合纸张。我会评论每个成员的所作所为。

　　点击打开代码下载链接

　　namespace cmt

{

class CMT

{

public:

CMT() : str_detector("FAST"), str_descriptor("BRISK") {} //默认的特征检测和描述子

void initialize(const Mat im_gray, const Rect rect); //rect是初始目标框

void processFrame(const Mat im_gray);

Fusion fusion;//跟踪点融合:融合跟踪和匹配的点 将两种点都放在一起，并且不重复

Matcher matcher; //DescriptorMatcher 用的knnMatch匹配

Tracker tracker; //光溜匹配跟踪 上一帧的特征点使用光流法跟踪得到这一帧的特征点的位置T

Consensus consensus;//包括scale和rotation angel的求取和对目标中心的投票

string str_detector;

string str_descriptor;

vector points_active; //public for visualization purposes

RotatedRect bb_rot;

private:

Ptr detector;

Ptr descriptor;

Size2f size_initial;

vector classes_active; //forgroud points -in target boundingboxs

float theta;

Mat im_prev;

};

} /* namespace CMT */

　　最重要的跟踪实施过程：

　　初始代码实现，添加注释：

　　void CMT::initialize(const Mat im_gray, const Rect rect)

{

//Remember initial size

size_initial = rect.size();

//Remember initial image

im_prev = im_gray;

//Compute center of rect

Point2f center = Point2f(rect.x + rect.width/2.0, rect.y + rect.height/2.0);

//Initialize rotated bounding box

bb_rot = RotatedRect(center, size_initial, 0.0);

//Initialize detector and descriptor

//FeatureDetector is OpenCVs feature detector,including "FAST","ORB","SIFT"

//"STAR",MSER","GFTT","HARRIS","Dense","SimpleBlob"

detector = FeatureDetector::create(str_detector);//FAST

descriptor = DescriptorExtractor::create(str_descriptor);//BRISK

//Get initial keypoints in whole image and compute their descriptors

vector keypoints;

detector->detect(im_gray, keypoints);

//Divide keypoints into foreground and background keypoints according to selection

//in target bounding box is foreground

vector keypoints_fg;

vector keypoints_bg;

for (size_t i = 0; i < keypoints.size(); i++)

{

KeyPoint k = keypoints[i];

Point2f pt = k.pt;

if (pt.x > rect.x && pt.y > rect.y && pt.x < rect.br().x && pt.y < rect.br().y)

{

keypoints_fg.push_back(k);

}

else

{

keypoints_bg.push_back(k);

}

}

//Create foreground classes

vector classes_fg;

classes_fg.reserve(keypoints_fg.size());

for (size_t i = 0; i < keypoints_fg.size(); i++)

{

classes_fg.push_back(i);

}

//Compute foreground/background features

Mat descs_fg;

Mat descs_bg;

descriptor->compute(im_gray, keypoints_fg, descs_fg);

descriptor->compute(im_gray, keypoints_bg, descs_bg);

//Only now is the right time to convert keypoints to points, as compute() might remove some keypoints

vector points_fg;

vector points_bg;

for (size_t i = 0; i < keypoints_fg.size(); i++)

{

points_fg.push_back(keypoints_fg[i].pt);

}

for (size_t i = 0; i < keypoints_bg.size(); i++)

{

points_bg.push_back(keypoints_bg[i].pt);

}

//Create normalized points: distance to center

vector points_normalized;

for (size_t i = 0; i < points_fg.size(); i++)

{

points_normalized.push_back(points_fg[i] - center);

}

//Initialize matcher

matcher.initialize(points_normalized, descs_fg, classes_fg, descs_bg, center);

//Initialize consensus get Xi,Xj distance and angle

consensus.initialize(points_normalized);

//Create initial set of active keypoints

for (size_t i = 0; i < keypoints_fg.size(); i++)

{

points_active.push_back(keypoints_fg[i].pt);

classes_active = classes_fg;

}

}

　　

　　总结：这是一种基于特征点匹配的跟踪算法。创新之处在于它利用特征点相对距离的投票来确定目标位置。优点是对于没有变形的物体，无论物体如何移动或旋转，上述特征点到中心的距离都是在缩放比例下确定的。

　　作者在论文中说：CMT背后的主要思想是将感兴趣的对象分解成微小的部分，称为关键点。在每一帧中，我们尝试再次找到已经在初始选择中的关键点感兴趣的对象。我们通过使用两种不同的方法来做到这一点。首先，我们通过估计所谓的光流来跟踪从前一帧到当前帧的关键点。其次，我们通过比较它们的描述符来全局匹配关键点。这些方法容易出错，我们采用了一种新颖的方式，通过让每个关键点投票给对象中心来在找到的关键点内寻找共识，如下图所示：

　　然后通过对中心集群进行投票来确定中心位置：