级联分类器由17个弱分类器串联组成，其中每个弱分类器关注一个特征。
The cascade consists of 17 weak classifiers, each concentrating on one feature.
计算样本图像的LBP直方图和模板的LBP直方图之间的度量，并由此构建弱分类器集。
The Chi square distance between corresponding LBP histograms of sample image and template is used to construct weak classifiers pool.
快速的人脸检测还有一个重要任务就是将这些弱分类器有效地组合起来，提高人脸检测的速度。
Most effectively combing the weak classifiers is important work for increasing the speed of detecting task.
基于委员会的方法试图通过合并多个弱分类器建立一个有效的委员会来构造一个更加有效的分类器。
The voting algorithms try to form a powerful classifier by combining multiple weak classifiers as a council of base-classifiers.
算法模拟弱分类器系数增加或减少对分类结果的影响，选择最利于分类的系数调整方向，设计了优化机制。
The algorithm simulated the effect of coefficient adjustment on the classifier result, chose the best direction of the classifier result and designed the optimization mechanism.
在该演化算法中，采取训练正反类样本加权模板的方法来构造各个弱学习分类器，克服了常规的基于单一特征构造弱分类器的不足。
The algorithm used weighted templates to structure each weak learning classifier, which overcame the shortcoming of structuring classifier by using a single feature.
对两种阈值和偏置计算方法的仿真实验结果表明，在错分率降可接受的范围内，二者均使用较少的弱分类器便可获得高识别率的强分类器。
Simulation experiments show when the error rate is in an acceptable range, the algorithms using fewer weak classifiers will be able to guarantee the strong classifier to maintain a high correct rate.
对两种阈值和偏置计算方法的仿真实验结果表明，在错分率降可接受的范围内，二者均使用较少的弱分类器便可获得高识别率的强分类器。
Simulation experiments show when the error rate is in an acceptable range, the algorithms using fewer weak classifiers will be able to guarantee the strong classifier to maintain a high correct rate.
应用推荐