| 嵌入式驾驶员状态检测算法的实现与优化 |
| |
| 引用本文： | 张旭,李亚利,陈晨,王生进,丁晓青.嵌入式驾驶员状态检测算法的实现与优化[J].自动化学报,2012,38(12):2014-2022. |
| |
| 作者姓名： | 张旭 李亚利 陈晨 王生进 丁晓青 |
| |
| 作者单位： | 1.智能技术与系统国家重点实验室 北京 100084, 中国; |
| |
| 基金项目： | 国家高技术研究发展计划(863计划)(2009AA11Z214);国家自然科学基金(61071135);教育部博士点基金项目(20090002110077)资助~~ |
| |
| 摘 要： | 提出了一种可以在嵌入式平台上实时运行的驾驶员状态检测算法. 状态检测采用了基于统计学习的Adaboost算法与动态建模算法. 与传统的采用主动红外光的方法相比, 本系统采用对人眼更为安全的被动式方法, 且对光线的变化有更好的鲁棒性. 算法的主要创新点是: 1) 提出了检测区域自适应调整的单双眼检测相结合的Adaboost人眼检测算法, 提高了人眼检测的准确性与速度; 2)提出基于高斯混合模型的人眼动态建模跟踪算法, 自动提取驾驶员眼睛区域灰度分布的信息, 实现了对不同驾驶员人眼的建模与跟踪定位. 在多个公共数据集以及实车采集的视频上进行的实验表明, 该算法能够准确判断驾驶员的状态, 满足实时处理的要求.
|
| 关 键 词： | 嵌入式系统 驾驶员状态检测 人眼动态模型 人眼跟踪 |
| 收稿时间： | 2012-01-04 |
Realization and Optimization of Embedded Driver Status Detection System |
| |
| ZHANG Xu,LI Ya-Li,CHEN Chen,WANG Sheng-Jin,DING Xiao-Qing.Realization and Optimization of Embedded Driver Status Detection System[J].Acta Automatica Sinica,2012,38(12):2014-2022. |
| |
| Authors: | ZHANG Xu LI Ya-Li CHEN Chen WANG Sheng-Jin DING Xiao-Qing |
| |
| Affiliation: | 1.State Key Laboratory of Intelligent Technology and Systems, Beijing 100084, China;2.Tsinghua National Laboratory for Information Science and Technology, Beijing 100084, China;3.Department of Electronic Engineering, Tsinghua University, Beijing 100084, China;4.Department of Computer &Information Science, University of Pennsylvania, Philadelphia PA 19104-6303, USA |
| |
| Abstract: | This paper presents a novel driver status detection algorithm, which can be processed in embedded platform in real-time. The proposed method is based on Adaboost and dynamic modeling algorithm. Compared to the traditional active infrared radiation method, our system employs a safer passive way and the algorithm is more robust to the various illuminations. There are two main contributions: 1) it mixes up the single eye and eye pairs detectors together and presents an adaptive detection region Adaboost eye detection algorithm, improving the detection rate and speeding up the eye detection process; 2) it presents a dynamic eye modeling tracking algorithm which is based on the Gaussian mixture model. The tracking algorithm can automatically extract image intensity distribution of the driver's eye region, thus can model and track eyes of different drivers. Experiments on several public databases and driving sequences taken in real car show that the proposed method can detect the driver status precisely and satisfy the real time processing requirements. |
| |
| Keywords: | Embedded system driver status detection dynamic eye model eye tracking |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《自动化学报》浏览原始摘要信息 |
|
点击此处可从《自动化学报》下载全文 |
|
