韩宪忠,李得锋,王克俭,周利亚.基于自编码神经网络与AdaBoost的快速行人检测算法[J].中南民族大学学报自然科学版,2018,(1):108-113
基于自编码神经网络与AdaBoost的快速行人检测算法
Fast Pedestrian Detection Algorithm Based on Auto-encoder Neural Network and AdaBoost
  
DOI:
中文关键词: 行人检测  HOG 特征  AdaBoost 算法  自编码网络  ACF 模型
英文关键词: pedestrian detection  HOG feature  AdaBoost algorithm  auto-encoder network  ACF model
基金项目:河北省科技项目基金资助项目(14227404D);河北农业大学理工基金项目(LG201407;ZD201407;LG20140703);河北省高等学校科学技术研究项目(ZD2015054)
作者单位
韩宪忠,李得锋,王克俭,周利亚 河北农业大学 信息科学与技术学院保定 071000 
摘要点击次数: 180
全文下载次数: 155
中文摘要:
      针对传统基于HOG特征与AdaBoost算法分类器在目标检测中存在检测速度慢、误差率大的问题,提出了一种基于自编码神经网络与AdaBoost的快速行人检测算法.该算法首先利用基于ACF模型的目标检测算法对图像进行预处理,获得疑似目标区域;然后对获取的子区域进行尺度归一化,提取HOG特征,并输入到自编码神经网络中进行降维;最后利用 AdaBoost 分类器对分类检测,输出检测到的行人区域.实验结果表明:文中所提算法的行人检测性能超过现有的检测算法,其检测速度也超过大多数算法.
英文摘要:
      Since the traditional algorithm has the shortages of slow detection rate and large error rate in pedestrian detection, a fast pedestrian detection algorithm based on auto-encoder neural network and AdaBoost is proposed. Firstly,the pedestrian detection algorithm based on ACF model is used to process the image to obtain the suspected object area.Then the acquired sub-region is normalized and the HOG feature is extracted and input into the auto-encoder neural network. Finally, the AdaBoost classifier is used to detect the classification and output the detected pedestrian area. The experimental results show that the proposed method has more performance than the existing detection algorithm for pedestrian detection, and its detection speed is also faster than most of the algorithms.
查看全文   查看/发表评论  下载PDF阅读器
关闭