一种基于改进Adaboost的人体检测算法

本文采用改进的Adaboost算法对静态图像中人体进行检测。针对传统算法中训练速度较慢，并且存在风险敏感的问题。本文提出一种快速特征选择算法，通过构造统计表，保存特征信息，避免每轮弱分类器训练时对所有特征重新计算分类误差；并结合fisher判别分析对选出的弱分类器进行训练，学习得到一个新的线性判别方程，最大化不同类别数据之间的可分离性，达到优化强分类器降低风险敏感影响的目的。实验结果表明，相对于传统的Adaboost算法，本文给出的方法加快了特征选择的速度，并具有较好的检测性能。

A Human Detection Algorithm Based on Improved Adaboost

This paper adopts an improved Adaboost algorithm to detect human in still images. Due to the slow training speed and the cost sensitive problem in traditional algorithm,this paper proposes a quick feature selecting algorithm to avoid calculating the classification error in the iteration,by means of establishing a structure statistics table to store the feature information. In the same time combining with fisher discriminant analysis,we optimize the weights of weak learner,learn to acquire a new linear discr...