| 基于改进的AdaBoost和支持向量机的行人检测 |
| |
| 引用本文： | 周维柏,李蓉.基于改进的AdaBoost和支持向量机的行人检测[J].昆明理工大学学报(理工版),2010,35(6):61-66. |
| |
| 作者姓名： | 周维柏 李蓉 |
| |
| 作者单位： | 华南师范大学增城学院,广东广州511363 |
| |
| 基金项目： | 广东省本科高等教育教学改革项目 |
| |
| 摘 要： | 针对AdaBoost算法在训练样本和特征较多时训练时间过长的问题,提出了一种改进的AdaBoot算法与支持向量机组合的分类器.对多重分类器的输出结果以非线性的方式组合,采用交替的方式轮流对不同的特征进行学习,将多重分类器处理完后的结果作为另一种输入样本,再以一个分类器做一次分类.实验表明该算法用于行人检测可行、性能稳定.
|
| 关 键 词： | AdaBoost 支持向量机 多重分类器 行人检测 |
Pedestrian Detection Based on Improved Adaboost and SVM Algorithm |
| |
| ZHOU Wei-bai,LI Rong.Pedestrian Detection Based on Improved Adaboost and SVM Algorithm[J].Journal of Kunming University of Science and Technology(Natural Science Edition),2010,35(6):61-66. |
| |
| Authors: | ZHOU Wei-bai LI Rong |
| |
| Affiliation: | (Zengcheng College,South China Normal University,Guangzhou 511363,China) |
| |
| Abstract: | A combined Classifier based on improved AdaBoost and Support Vector Machine,is proposed in order to deal with the problems of long training time when the number of training samples and features increase.It can combine multiple classifiers in a nonlinear manner.Different characteristics are studied in Alternate way by turns.The results of multiple classifiers serve as another input samples.Then a classifier is to do a more classification.The experiment result indicates that the algorithm is fast and reliable in both training and pedestrian detection. |
| |
| Keywords: | AdaBoost Support Vector Machine（SVM） multiple classifier pedestrian detection |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
