人脸识别之人脸检测（九）--检测器源码分析

来源：互联网发布：h5 css3网站模版源码编辑：程序博客网时间：2024/05/16 04:53

原文：

http://blog.csdn.net/xidianzhimeng/article/details/41851569

还有几篇蛮好：

http://blog.csdn.net/delltdk/article/details/23739441 总体描述，总结。

http://blog.csdn.net/tracyliang223/article/details/17379769，以HOG为例解释。

http://blog.csdn.net/naruto0001/article/details/8086745，以LBP为例解释。

http://blog.csdn.net/beerbuddys/article/details/40712957，以HARR为例解释。

级联分类器检测类CascadeClassifier，在2.4.5版本中使用Adaboost的方法+LBP、HOG、HAAR进行目标检测，加载的是使用traincascade进行训练的分类器

class CV_EXPORTS_W CascadeClassifier

{

public:

CV_WRAP CascadeClassifier(); // 无参数构造函数，new自动调用该函数分配初试内存

CV_WRAP CascadeClassifier( const string& filename ); // 带参数构造函数，参数为XML的绝对名称

virtual ~CascadeClassifier(); // 析构函数，无需关心

CV_WRAP virtual bool empty() const; // 是否导入参数，只创建了该对象而没有加载或者加载失败时都是空的

CV_WRAP bool load( const string& filename ); // 加载分类器，参数为XML的绝对名称，函数内部调用read读取新格式的分类器，读取成功后直接返回，读取失败后调用cvLoad读取旧格式的分类器，读取成功返回true，否则返回false

virtual bool read( const FileNode& node ); // load内部调用read解析XML中的内容，也可以自己创建节点然后调用Read即可，但是该函数只能读取新格式的分类器，不能读取旧格式的分类器

// 多尺度检测函数

CV_WRAP virtual void detectMultiScale( const Mat& image, // 图像，cvarrtoMat实现IplImage转换为Mat，必须为8位，内部可自行转换为灰度图像

CV_OUT vector<Rect>& objects, // 输出矩形，注意vector不是线程安全的

double scaleFactor=1.1, // 缩放比例，必须大于1

int minNeighbors=3, // 合并窗口时最小neighbor，每个候选矩阵至少包含的附近元素个数

int flags=0, // 检测标记，只对旧格式的分类器有效，与cvHaarDetectObjects的参数flags相同，默认为0，可能的取值为CV_HAAR_DO_CANNY_PRUNING(CANNY边缘检测)、CV_HAAR_SCALE_IMAGE(缩放图像)、CV_HAAR_FIND_BIGGEST_OBJECT(寻找最大的目标)、CV_HAAR_DO_ROUGH_SEARCH(做粗略搜索)；如果寻找最大的目标就不能缩放图像，也不能CANNY边缘检测

Size minSize=Size(), // 最小检测目标

Size maxSize=Size() ); // 最大检测目标

// 最好不要在这里设置最大最小，可能会影响合并的效果，因此可以在检测完毕后自行判断结果是否满足要求

CV_WRAP virtual void detectMultiScale( const Mat& image,

CV_OUT vector<Rect>& objects,

vector<int>& rejectLevels,

vector<double>& levelWeights,

double scaleFactor=1.1,

int minNeighbors=3, int flags=0,

Size minSize=Size(),

Size maxSize=Size(),

bool outputRejectLevels=false );

// 上述参数多了rejectLevels和levelWeights以及outputRejectLevels参数，只有在outputRejectLevels为true的时候才可能输出前两个参数

// 还有就是在使用旧分类器的时候必须设置flags为CV_HAAR_SCALE_IMAGE，可以通过haarcascade_frontalface_alt.xml检测人脸尝试

bool isOldFormatCascade() const; // 是否是旧格式的分类器

virtual Size getOriginalWindowSize() const; // 初始检测窗口大小，也就是训练的窗口

int getFeatureType() const; // 获取特征类型

bool setImage( const Mat& ); // 设置图像，计算图像的积分图

virtual int runAt( Ptr<FeatureEvaluator>& feval, Point pt, double& weight ); // 计算某检测窗口是否为目标

// 保存强分类器数据

class Data

{

public:

struct CV_EXPORTS DTreeNode // 节点

{

int featureIdx; // 对应的特征编号

float threshold; // for ordered features only 节点阈值

int left; // 左子树

int right; // 右子树

};

struct CV_EXPORTS DTree // 弱分类器

{

int nodeCount; // 弱分类器中节点个数

};

struct CV_EXPORTS Stage // 强分类器

{

int first; // 在classifier中的起始位置

int ntrees; // 该强分类器中的弱分类器数

float threshold; // 强分类器阈值

};

bool read(const FileNode &node); // 读取强分类器

bool isStumpBased; // 是否只有树桩

int stageType; // BOOST，boostType:GAB、RAB等

int featureType; // HAAR、HOG、LBP

int ncategories; // maxCatCount，LBP为256，其余为0

Size origWinSize;

vector<Stage> stages;

vector<DTree> classifiers;

vector<DTreeNode> nodes;

vector<float> leaves;

vector<int> subsets;

};

Data data;

Ptr<FeatureEvaluator> featureEvaluator;

Ptr<CvHaarClassifierCascade> oldCascade;

// 关于mask这块参考《OpenCV目标检测之MaskGenerator》

public:

class CV_EXPORTS MaskGenerator

{

public:

virtual ~MaskGenerator() {}

virtual cv::Mat generateMask(const cv::Mat& src)=0;

virtual void initializeMask(const cv::Mat& /*src*/) {};

};

void setMaskGenerator(Ptr<MaskGenerator> maskGenerator);

Ptr<MaskGenerator> getMaskGenerator();

void setFaceDetectionMaskGenerator();

protected:

Ptr<MaskGenerator> maskGenerator;

}

注意：当在不同的分类器之间切换的时候，需要手动释放，因为read内部没有释放上一次读取的分类器数据！

关于新旧格式的分类器参考《OpenCV存储解读之Adaboost分类器》

使用CascadeClassifier检测目标的过程

1) load分类器并调用empty函数检测是否load成功

// 读取stages

bool CascadeClassifier::Data::read(const FileNode &root)

{

static const float THRESHOLD_EPS = 1e-5f;

// load stage params

string stageTypeStr = (string)root[CC_STAGE_TYPE];

if( stageTypeStr == CC_BOOST )

stageType = BOOST;

else

return false;

printf("stageType: %s\n", stageTypeStr.c_str());

string featureTypeStr = (string)root[CC_FEATURE_TYPE];

if( featureTypeStr == CC_HAAR )

featureType = FeatureEvaluator::HAAR;

else if( featureTypeStr == CC_LBP )

featureType = FeatureEvaluator::LBP;

else if( featureTypeStr == CC_HOG )

featureType = FeatureEvaluator::HOG;

else

return false;

printf("featureType: %s\n", featureTypeStr.c_str());

origWinSize.width = (int)root[CC_WIDTH];

origWinSize.height = (int)root[CC_HEIGHT];

CV_Assert( origWinSize.height > 0 && origWinSize.width > 0 );

isStumpBased = (int)(root[CC_STAGE_PARAMS][CC_MAX_DEPTH]) == 1 ? true : false;

printf("stumpBased: %d\n", isStumpBased);

// load feature params

FileNode fn = root[CC_FEATURE_PARAMS];

if( fn.empty() )

return false;

// LBP的maxCatCount=256，其余特征都等于0

ncategories = fn[CC_MAX_CAT_COUNT]; // ncategories=256/0

int subsetSize = (ncategories + 31)/32,// subsetSize=8/0 // 强制类型转换取整，不是四舍五入

nodeStep = 3 + ( ncategories>0 ? subsetSize : 1 ); //每组数值个数，nodeStep=11/4

printf("subsetSize: %d, nodeStep: %d\n", subsetSize, nodeStep);

// load stages

fn = root[CC_STAGES];

if( fn.empty() )

return false;

stages.reserve(fn.size());

classifiers.clear();

nodes.clear();

FileNodeIterator it = fn.begin(), it_end = fn.end();

for( int si = 0; it != it_end; si++, ++it )

{

FileNode fns = *it;

Stage stage;

stage.threshold = (float)fns[CC_STAGE_THRESHOLD] - THRESHOLD_EPS;

fns = fns[CC_WEAK_CLASSIFIERS];

if(fns.empty())

return false;

stage.ntrees = (int)fns.size();

stage.first = (int)classifiers.size();

printf("stage %d: ntrees: %d, first: %d\n", si, stage.ntrees, stage.first);

stages.push_back(stage);

classifiers.reserve(stages[si].first + stages[si].ntrees);

FileNodeIterator it1 = fns.begin(), it1_end = fns.end();

for( ; it1 != it1_end; ++it1 ) // weak trees

{

FileNode fnw = *it1;

FileNode internalNodes = fnw[CC_INTERNAL_NODES];

FileNode leafValues = fnw[CC_LEAF_VALUES];

if( internalNodes.empty() || leafValues.empty() )

return false;

// 弱分类器中的节点

DTree tree;

tree.nodeCount = (int)internalNodes.size()/nodeStep;

classifiers.push_back(tree);

nodes.reserve(nodes.size() + tree.nodeCount);

leaves.reserve(leaves.size() + leafValues.size());

if( subsetSize > 0 ) // 针对LBP

subsets.reserve(subsets.size() + tree.nodeCount*subsetSize);

FileNodeIterator internalNodesIter = internalNodes.begin(), internalNodesEnd = internalNodes.end();

// 保存每一个node

for( ; internalNodesIter != internalNodesEnd; ) // nodes

{

DTreeNode node;

node.left = (int)*internalNodesIter; ++internalNodesIter;

node.right = (int)*internalNodesIter; ++internalNodesIter;

node.featureIdx = (int)*internalNodesIter; ++internalNodesIter;

// 针对LBP，获取8个数值

if( subsetSize > 0 )

{

for( int j = 0; j < subsetSize; j++, ++internalNodesIter )

subsets.push_back((int)*internalNodesIter);

node.threshold = 0.f;

}

else

{

node.threshold = (float)*internalNodesIter; ++internalNodesIter;

}

nodes.push_back(node);

}

// 保存叶子节点

internalNodesIter = leafValues.begin(), internalNodesEnd = leafValues.end();

for( ; internalNodesIter != internalNodesEnd; ++internalNodesIter ) // leaves

leaves.push_back((float)*internalNodesIter);

}

return true;

}

// 读取stages与features

bool CascadeClassifier::read(const FileNode& root)

{

// load stages

if( !data.read(root) )

return false;

// load features，参考《图像特征->XXX特征之OpenCV-估计》

featureEvaluator = FeatureEvaluator::create(data.featureType);

FileNode fn = root[CC_FEATURES];

if( fn.empty() )

return false;

return featureEvaluator->read(fn);

}

// 外部调用的函数

bool CascadeClassifier::load(const string& filename)

{

oldCascade.release();

data = Data();

featureEvaluator.release();

// 读取新格式的分类器

FileStorage fs(filename, FileStorage::READ);

if( !fs.isOpened() )

return false;

if( read(fs.getFirstTopLevelNode()) )

return true;

fs.release();

// 读取新格式失败则读取旧格式的分类器

oldCascade = Ptr<CvHaarClassifierCascade>((CvHaarClassifierCascade*)cvLoad(filename.c_str(), 0, 0, 0));

return !oldCascade.empty();

}

2) 调用detectMultiScale函数进行多尺度检测，该函数可以使用老分类器进行检测也可以使用新分类器进行检测

2.1 如果load的为旧格式的分类器则使用cvHaarDetectObjectsForROC进行检测，flags参数只对旧格式的分类器有效，参考《OpenCV函数解读之cvHaarDetectObjects》

if( isOldFormatCascade() )

{

MemStorage storage(cvCreateMemStorage(0));

CvMat _image = image;

CvSeq* _objects = cvHaarDetectObjectsForROC( &_image, oldCascade, storage, rejectLevels, levelWeights, scaleFactor,

minNeighbors, flags, minObjectSize, maxObjectSize, outputRejectLevels );

vector<CvAvgComp> vecAvgComp;

Seq<CvAvgComp>(_objects).copyTo(vecAvgComp);

objects.resize(vecAvgComp.size());

std::transform(vecAvgComp.begin(), vecAvgComp.end(), objects.begin(), getRect());

return;

}

2.2 新格式分类器多尺度检测

for( double factor = 1; ; factor *= scaleFactor )

{

Size originalWindowSize = getOriginalWindowSize();

Size windowSize( cvRound(originalWindowSize.width*factor), cvRound(originalWindowSize.height*factor) );

Size scaledImageSize( cvRound( grayImage.cols/factor ), cvRound( grayImage.rows/factor ) );

Size processingRectSize( scaledImageSize.width-originalWindowSize.width + 1, scaledImageSize.height-originalWindowSize.height + 1 );

if( processingRectSize.width <= 0 || processingRectSize.height <= 0 )

break;

if( windowSize.width > maxObjectSize.width || windowSize.height > maxObjectSize.height )

break;

if( windowSize.width < minObjectSize.width || windowSize.height < minObjectSize.height )

continue;

// 缩放图像

Mat scaledImage( scaledImageSize, CV_8U, imageBuffer.data );

resize( grayImage, scaledImage, scaledImageSize, 0, 0, CV_INTER_LINEAR );

// 计算步长

int yStep;

if( getFeatureType() == cv::FeatureEvaluator::HOG )

{

yStep = 4;

}

else

{

yStep = factor > 2. ? 1 : 2;

}

// 并行个数以及大小，按照列进行并行处理

int stripCount, stripSize;

// 是否采用TBB进行优化

#ifdef HAVE_TBB

const int PTS_PER_THREAD = 1000;

stripCount = ((processingRectSize.width/yStep)*(processingRectSize.height + yStep-1)/yStep + PTS_PER_THREAD/2)/PTS_PER_THREAD;

stripCount = std::min(std::max(stripCount, 1), 100);

stripSize = (((processingRectSize.height + stripCount - 1)/stripCount + yStep-1)/yStep)*yStep;

#else

stripCount = 1;

stripSize = processingRectSize.height;

#endif

// 调用单尺度检测函数进行检测

if( !detectSingleScale( scaledImage, stripCount, processingRectSize, stripSize, yStep, factor, candidates,

rejectLevels, levelWeights, outputRejectLevels ) )

break;

}

2.3 合并检测结果

objects.resize(candidates.size());

std::copy(candidates.begin(), candidates.end(), objects.begin());

if( outputRejectLevels )

{

groupRectangles( objects, rejectLevels, levelWeights, minNeighbors, GROUP_EPS );

}

else

{

groupRectangles( objects, minNeighbors, GROUP_EPS );

}

单尺度检测函数流程

2.2.1 根据所载入的特征计算积分图、积分直方图等

// 计算当前图像的积分图，参考《图像特征->XXX特征之OpenCV-估计》

if( !featureEvaluator->setImage( image, data.origWinSize ) )

return false;

2.2.2 根据是否输出检测级数并行目标检测

vector<Rect> candidatesVector;

vector<int> rejectLevels;

vector<double> levelWeights;

Mutex mtx;

if( outputRejectLevels )

{

parallel_for_(Range(0, stripCount), CascadeClassifierInvoker( *this, processingRectSize, stripSize, yStep, factor,

candidatesVector, rejectLevels, levelWeights, true, currentMask, &mtx));

levels.insert( levels.end(), rejectLevels.begin(), rejectLevels.end() );

weights.insert( weights.end(), levelWeights.begin(), levelWeights.end() );

}

else

{

parallel_for_(Range(0, stripCount), CascadeClassifierInvoker( *this, processingRectSize, stripSize, yStep, factor,

candidatesVector, rejectLevels, levelWeights, false, currentMask, &mtx));

}

candidates.insert( candidates.end(), candidatesVector.begin(), candidatesVector.end() );

CascadeClassifierInvoker函数的operator()实现具体的检测过程

// 对于没有并行时range.start=0,range.end=1

void operator()(const Range& range) const

{

Ptr<FeatureEvaluator> evaluator = classifier->featureEvaluator->clone();

Size winSize(cvRound(classifier->data.origWinSize.width * scalingFactor),

cvRound(classifier->data.origWinSize.height * scalingFactor));

// strip=processingRectSize.height

int y1 = range.start * stripSize; // 0

int y2 = min(range.end * stripSize, processingRectSize.height); // processSizeRect.height也就是可以处理的高度，已经减去窗口高度

for( int y = y1; y < y2; y += yStep )

{

for( int x = 0; x < processingRectSize.width; x += yStep )

{

if ( (!mask.empty()) && (mask.at<uchar>(Point(x,y))==0)) {

continue;

}

// result=1表示通过了所有的分类器 <=0表示失败的级数

// gypWeight表示返回的阈值

double gypWeight;

int result = classifier->runAt(evaluator, Point(x, y), gypWeight);

// 输出LOG

#if defined (LOG_CASCADE_STATISTIC)

logger.setPoint(Point(x, y), result);

#endif

// 当返回级数的时候可以最后三个分类器不通过

if( rejectLevels )

{

if( result == 1 )

result = -(int)classifier->data.stages.size();

// 可以最后三个分类器不通过

if( classifier->data.stages.size() + result < 4 )

{

mtx->lock();

rectangles->push_back(Rect(cvRound(x*scalingFactor), cvRound(y*scalingFactor), winSize.width, winSize.height));

rejectLevels->push_back(-result);

levelWeights->push_back(gypWeight);

mtx->unlock();

}

// 不返回级数的时候通过所有的分类器才保存起来

else if( result > 0 )

{

mtx->lock();

rectangles->push_back(Rect(cvRound(x*scalingFactor), cvRound(y*scalingFactor),

winSize.width, winSize.height));

mtx->unlock();

}

// 如果一级都没有通过那么加大搜索步长

if( result == 0 )

x += yStep;

}

runAt函数实现某一检测窗口的检测

int CascadeClassifier::runAt( Ptr<FeatureEvaluator>& evaluator, Point pt, double& weight )

{

CV_Assert( oldCascade.empty() );

assert( data.featureType == FeatureEvaluator::HAAR ||

data.featureType == FeatureEvaluator::LBP ||

data.featureType == FeatureEvaluator::HOG );

// 设置某一点处的特征，参考《图像特征->XXX特征之OpenCV-估计》

if( !evaluator->setWindow(pt) )

return -1;

// 如果为树桩，没有树枝

if( data.isStumpBased )

{

if( data.featureType == FeatureEvaluator::HAAR )

return predictOrderedStump<HaarEvaluator>( *this, evaluator, weight );

else if( data.featureType == FeatureEvaluator::LBP )

return predictCategoricalStump<LBPEvaluator>( *this, evaluator, weight );

else if( data.featureType == FeatureEvaluator::HOG )

return predictOrderedStump<HOGEvaluator>( *this, evaluator, weight );

else

return -2;

}

// 每个弱分类器不止一个node

else

{

if( data.featureType == FeatureEvaluator::HAAR )

return predictOrdered<HaarEvaluator>( *this, evaluator, weight );

else if( data.featureType == FeatureEvaluator::LBP )

return predictCategorical<LBPEvaluator>( *this, evaluator, weight );

else if( data.featureType == FeatureEvaluator::HOG )

return predictOrdered<HOGEvaluator>( *this, evaluator, weight );

else

return -2;

}

predictOrdered*函数实现判断当前检测窗口的判断

// HAAR与HOG特征的多node检测

template<class FEval>

inline int predictOrdered( CascadeClassifier& cascade, Ptr<FeatureEvaluator> &_featureEvaluator, double& sum )

{

int nstages = (int)cascade.data.stages.size();

int nodeOfs = 0, leafOfs = 0;

FEval& featureEvaluator = (FEval&)*_featureEvaluator;

float* cascadeLeaves = &cascade.data.leaves[0];

CascadeClassifier::Data::DTreeNode* cascadeNodes = &cascade.data.nodes[0];

CascadeClassifier::Data::DTree* cascadeWeaks = &cascade.data.classifiers[0];

CascadeClassifier::Data::Stage* cascadeStages = &cascade.data.stages[0];

// 遍历每个强分类器

for( int si = 0; si < nstages; si++ )

{

CascadeClassifier::Data::Stage& stage = cascadeStages[si];

int wi, ntrees = stage.ntrees;

sum = 0;

// 遍历每个弱分类器

for( wi = 0; wi < ntrees; wi++ )

{

CascadeClassifier::Data::DTree& weak = cascadeWeaks[stage.first + wi];

int idx = 0, root = nodeOfs;

// 遍历每个节点

{

// 选择一个node：root和idx初始化为0，即第一个node

CascadeClassifier::Data::DTreeNode& node = cascadeNodes[root + idx];

// 计算当前node特征池编号下的特征值

double val = featureEvaluator(node.featureIdx);

// 如果val小于node阈值则选择左子树，否则选择右子树

idx = val < node.threshold ? node.left : node.right;

} while( idx > 0 );

// 累加最终的叶子节点

sum += cascadeLeaves[leafOfs - idx];

nodeOfs += weak.nodeCount;

leafOfs += weak.nodeCount + 1;

}

// 判断所有叶子节点累加和是否小于强分类器阈值，小于强分类器阈值则失败

if( sum < stage.threshold )

return -si;

}

// 通过了所有的强分类器返回1，否则返回失败的分类器

return 1;

}

// LBP特征的多node检测

template<class FEval>

inline int predictCategorical( CascadeClassifier& cascade, Ptr<FeatureEvaluator> &_featureEvaluator, double& sum )

{

int nstages = (int)cascade.data.stages.size();

int nodeOfs = 0, leafOfs = 0;

FEval& featureEvaluator = (FEval&)*_featureEvaluator;

size_t subsetSize = (cascade.data.ncategories + 31)/32;

int* cascadeSubsets = &cascade.data.subsets[0];

float* cascadeLeaves = &cascade.data.leaves[0];

CascadeClassifier::Data::DTreeNode* cascadeNodes = &cascade.data.nodes[0];

CascadeClassifier::Data::DTree* cascadeWeaks = &cascade.data.classifiers[0];

CascadeClassifier::Data::Stage* cascadeStages = &cascade.data.stages[0];

for(int si = 0; si < nstages; si++ )

{

CascadeClassifier::Data::Stage& stage = cascadeStages[si];

int wi, ntrees = stage.ntrees;

sum = 0;

for( wi = 0; wi < ntrees; wi++ )

{

CascadeClassifier::Data::DTree& weak = cascadeWeaks[stage.first + wi];

int idx = 0, root = nodeOfs;

{

CascadeClassifier::Data::DTreeNode& node = cascadeNodes[root + idx];

// c为0-255之间的数

int c = featureEvaluator(node.featureIdx);

// 获取当前node的subset头位置

const int* subset = &cascadeSubsets[(root + idx)*subsetSize]; // LBP:subsetSize=8

// 判断选择左子树还是右子树

idx = (subset[c>>5] & (1 << (c & 31))) ? node.left : node.right;

// c>>5表示将c右移5位，选择高3位，0-7之间

// c&31表示低5位，1<<(c&31)选择低5位后左移1位

// 将上面的数按位与，如果最后结果不为0表示选择左子树，否则选择右子树

}

while( idx > 0 );

sum += cascadeLeaves[leafOfs - idx];

nodeOfs += weak.nodeCount;

leafOfs += weak.nodeCount + 1;

}

if( sum < stage.threshold )

return -si;

}

return 1;

}

// HAAR与HOG特征的单node检测

template<class FEval>

inline int predictOrderedStump( CascadeClassifier& cascade, Ptr<FeatureEvaluator> &_featureEvaluator, double& sum )

{

int nodeOfs = 0, leafOfs = 0;

FEval& featureEvaluator = (FEval&)*_featureEvaluator;

float* cascadeLeaves = &cascade.data.leaves[0];

CascadeClassifier::Data::DTreeNode* cascadeNodes = &cascade.data.nodes[0];

CascadeClassifier::Data::Stage* cascadeStages = &cascade.data.stages[0];

int nstages = (int)cascade.data.stages.size();

// 遍历每个强分类器

for( int stageIdx = 0; stageIdx < nstages; stageIdx++ )

{

CascadeClassifier::Data::Stage& stage = cascadeStages[stageIdx];

sum = 0.0;

int ntrees = stage.ntrees;

// 遍历每个弱分类器

for( int i = 0; i < ntrees; i++, nodeOfs++, leafOfs+= 2 )

{

CascadeClassifier::Data::DTreeNode& node = cascadeNodes[nodeOfs];

double value = featureEvaluator(node.featureIdx);

sum += cascadeLeaves[ value < node.threshold ? leafOfs : leafOfs + 1 ];

}

if( sum < stage.threshold )

return -stageIdx;

}

return 1;

}

// LBP特征的单node检测

template<class FEval>

inline int predictCategoricalStump( CascadeClassifier& cascade, Ptr<FeatureEvaluator> &_featureEvaluator, double& sum )

{

int nstages = (int)cascade.data.stages.size();

int nodeOfs = 0, leafOfs = 0;

FEval& featureEvaluator = (FEval&)*_featureEvaluator;

size_t subsetSize = (cascade.data.ncategories + 31)/32;

int* cascadeSubsets = &cascade.data.subsets[0];

float* cascadeLeaves = &cascade.data.leaves[0];

CascadeClassifier::Data::DTreeNode* cascadeNodes = &cascade.data.nodes[0];

CascadeClassifier::Data::Stage* cascadeStages = &cascade.data.stages[0];

#ifdef HAVE_TEGRA_OPTIMIZATION

float tmp = 0; // float accumulator -- float operations are quicker

#endif

for( int si = 0; si < nstages; si++ )

{

CascadeClassifier::Data::Stage& stage = cascadeStages[si];

int wi, ntrees = stage.ntrees;

#ifdef HAVE_TEGRA_OPTIMIZATION

tmp = 0;

#else

sum = 0;

#endif

for( wi = 0; wi < ntrees; wi++ )

{

CascadeClassifier::Data::DTreeNode& node = cascadeNodes[nodeOfs];

int c = featureEvaluator(node.featureIdx);

const int* subset = &cascadeSubsets[nodeOfs*subsetSize];

#ifdef HAVE_TEGRA_OPTIMIZATION

tmp += cascadeLeaves[ subset[c>>5] & (1 << (c & 31)) ? leafOfs : leafOfs+1];

#else

sum += cascadeLeaves[ subset[c>>5] & (1 << (c & 31)) ? leafOfs : leafOfs+1];

#endif

nodeOfs++;

leafOfs += 2;

}

#ifdef HAVE_TEGRA_OPTIMIZATION

if( tmp < stage.threshold ) {

sum = (double)tmp;

return -si;

}

#else

if( sum < stage.threshold )

return -si;

#endif

}

#ifdef HAVE_TEGRA_OPTIMIZATION

sum = (double)tmp;

#endif

return 1;

}

0 0