摘要
基于AdaBoost分类器的图像/视频目标检测系统具有检测精度高、检测速度快的特点,但当训练样本数目多、样本描述的特征维数高时,分类器的训练过程将会异常缓慢。为有效改善分类器训练的时间性能,从限制弱分类器训练样本规模的角度,提出了一种改进的boosting分类器训练模型,即基于SC-AdaBoost的分类器训练模型。基于VOC2006数据集的车辆检测实验表明,在不损失分类器检测性能的前提下,SC-AdaBoost训练模型可明显减少分类器的训练时间。
Although AdaBoost-based object detection from image/video dada holds the characteristics of good detection precision and high detection speed,the training procedure is much more slowly especially when the number of both samples and feature dimensionality is high.With the aim of efficiently improving the training performance,this paper proposed an algorithm called SC-AdaBoost.Experimental results for car detection based on VOC2006 datasets show that when the number of training samples is very large,the proposed algorithm can evidently reduce the whole training time without loss of detection performance.
出处
《计算机科学》
CSCD
北大核心
2015年第7期309-313,共5页
Computer Science
基金
国家自然科学基金项目:动态纹理建模与应用的张量方法研究(11301137)
Spiking神经网络在移动机器人感知及控制中的应用研究(61175059)
无线移动智能视频监控系统中的数学方法(10926179)
河北省科技支撑计划项目:嵌入式校园视频监控系统(10243554D)
2012年河北师范大学应用开发基金资助
