The feature used in a particular classifier is specified by its shape (1a, 2b etc.), position within the region of interest and the scale (this scale is not the same as the scale used at the detection stage, though these two scales are multiplied). For example, in the case of the third line feature (2c) the response is calculated as the difference between the sum of image pixels under the rectangle covering the whole feature (including the two white stripes and the black stripe in the middle) and the sum of the image pixels under the black stripe multiplied by 3 in order to compensate for the differences in the size of areas. The sums of pixel values over a rectangular regions are calculated rapidly using integral images (see below and the integral description).

To see the object detector at work, have a look at the facedetect demo: https://github.com/Itseez/opencv/tree/master/samples/cpp/dbt_face_detection.cpp

The following reference is for the detection part only. There is a separate application called opencv_traincascade that can train a cascade of boosted classifiers from a set of samples.

In the new C++ interface it is also possible to use LBP (local binary pattern) features in addition to Haar-like features. .. [Viola01] Paul Viola and Michael J. Jones. Rapid Object Detection using a Boosted Cascade of Simple Features. IEEE CVPR, 2001. The paper is available online at https://www.cs.cmu.edu/~efros/courses/LBMV07/Papers/viola-cvpr-01.pdf（上述有提到）

7、opencv關于boost

(boost.cpp——opencv3.0)

Boosting

A common machine learning task is supervised learning. In supervised learning, the goal is to learn the functional relationship F:y=F(x) between the input x and the output y . Predicting the qualitative output is called classification, while predicting the quantitative output is called regression.

Boosting is a powerful learning concept that provides a solution to the supervised classification learning task. It combines the performance of many "weak" classifiers to produce a powerful committee [125] . A weak classifier is only required to be better than chance, and thus can be very simple and computationally inexpensive. However, many of them smartly combine results to a strong classifier that often outperforms most "monolithic" strong classifiers such as SVMs and Neural Networks.

Decision trees are the most popular weak classifiers used in boosting schemes. Often the simplest decision trees with only a single split node per tree (called stumps ) are sufficient.

The boosted model is based on N training examples (xi,yi)1N with xi∈RK and yi∈?1,+1 . xi is a K -component vector. Each component encodes a feature relevant to the learning task at hand. The desired two-class output is encoded as -1 and +1.

Different variants of boosting are known as Discrete Adaboost, Real AdaBoost, LogitBoost, and Gentle AdaBoost [49] . All of them are very similar in their overall structure. Therefore, this chapter focuses only on the standard two-class Discrete AdaBoost algorithm, outlined below. Initially the same weight is assigned to each sample (step 2). Then, a weak classifier fm(x) is trained on the weighted training data (step 3a). Its weighted training error and scaling factor cm is computed (step 3b). The weights are increased for training samples that have been misclassified (step 3c). All weights are then normalized, and the process of finding the next weak classifier continues for another M -1 times. The final classifier F(x) is the sign of the weighted sum over the individual weak classifiers (step 4).

Two-class Discrete AdaBoost Algorithm

To reduce computation time for boosted models without substantially losing accuracy, the influence trimming technique can be employed. As the training algorithm proceeds and the number of trees in the ensemble is increased, a larger number of the training samples are classified correctly and with increasing confidence, thereby those samples receive smaller weights on the subsequent iterations. Examples with a very low relative weight have a small impact on the weak classifier training. Thus, such examples may be excluded during the weak classifier training without having much effect on the induced classifier. This process is controlled with the weight_trim_rate parameter. Only examples with the summary fraction weight_trim_rate of the total weight mass are used in the weak classifier training. Note that the weights for all training examples are recomputed at each training iteration. Examples deleted at a particular iteration may be used again for learning some of the weak classifiers further [49]

Prediction with Boost

StatModel::predict(samples, results, flags) should be used. Pass flags=StatModel::RAW_OUTPUT to get the raw sum from Boost classifier.

8、關于訓練過程打印信息的解釋

1）POS count : consumed   n1 : n2

每次都調用updateTrainingSet( requiredLeafFARate, tempLeafFARate );函數

int CvCascadeClassifier::fillPassedSamples( int first, int count, bool isPositive, double minimumAcceptanceRatio, int64& consumed )  {      int getcount = 0;      Mat img(cascadeParams.winSize, CV_8UC1);      for( int i = first; i < first + count; i++ )      {          for( ; ; )          {              if( consumed != 0 && ((double)getcount+1)/(double)(int64)consumed <= minimumAcceptanceRatio )                  return getcount;                bool isGetImg = isPositive ? imgReader.getPos( img ) :                                             imgReader.getNeg( img );              if( !isGetImg )                  return getcount;              consumed++;                featureEvaluator->setImage( img, isPositive ? 1 : 0, i );              if( predict( i ) == 1.0F )              {                  getcount++;                  printf("%s current samples: %d/r", isPositive ? "POS":"NEG", getcount);                  break;              }          }      }      return getcount;  }  [cpp] view%20plain copy int CvCascadeClassifier::predict( int sampleIdx )  {      CV_DbgAssert( sampleIdx < numPos + numNeg );      for (vector< Ptr<CvCascadeBoost> >::iterator it = stageClassifiers.begin();          it != stageClassifiers.end(); it++ )      {          if ( (*it)->predict( sampleIdx ) == 0.f )              return 0;      }      return 1;  }  [cpp] view%20plain copy float CvCascadeBoost::predict( int sampleIdx, bool returnSum ) const  {      CV_Assert( weak );      double sum = 0;      CvSeqReader reader;      cvStartReadSeq( weak, &reader );      cvSetSeqReaderPos( &reader, 0 );      for( int i = 0; i < weak->total; i++ )      {          CvBoostTree* wtree;          CV_READ_SEQ_ELEM( wtree, reader );          sum += ((CvCascadeBoostTree*)wtree)->predict(sampleIdx)->value;      }      if( !returnSum )          sum = sum < threshold - CV_THRESHOLD_EPS ? 0.0 : 1.0;      return (float)sum;  }