基于GA-BP的安全带佩戴识别方法

发布时间：2018-01-13 03:09

本文关键词：基于GA-BP的安全带佩戴识别方法　出处：《江苏大学学报(自然科学版)》2014年02期 　论文类型：期刊论文

【摘要】：为了避免不规范佩戴安全带行为的发生,进一步提高安全带的佩戴率,提出了一种基于GA-BP的安全带佩戴识别方法.该方法在图像处理技术的基础上,提取安全带极坐标转化后的二值化图像像素值作为表征安全带佩戴状态的特征向量,并通过PCA方法对其进行降维;然后采用BP神经网络算法,建立基于BP神经网络的安全带佩戴识别模型,同时为了提高安全带佩戴识别模型的精度,引入遗传算法对其权值和阈值进行优化,建立基于GA-BP的安全带佩戴识别模型;最后通过具体实例验证.结果表明:该方法合理有效,能较好地对安全带的不同佩戴状态进行识别,具有较好的实用性和推广性.
[Abstract]:In order to avoid the occurrence of irregular wear safety belt behavior, to further improve the safety belt wearing rate, we propose a GA-BP recognition method based on seat belt wearing. This method is based on image processing technology, extracting the safety belt of polar coordinate value of two converted image pixel values as feature vectors for characterization of safety belt wearing state, and the reduction by PCA method; then using BP neural network algorithm, the establishment of safety belt wearing recognition model based on BP neural network, and in order to improve the accuracy of safety belt wearing recognition model, genetic algorithm is introduced to optimize the weights and thresholds, establish safety wearing recognition model based on GA-BP; finally verified by concrete examples. The results show that this method is reasonable and effective, to identify the different wear state can be better for the safety belt, has good practicability and validity.

【作者单位】：江苏大学汽车与交通工程学院;
【基金】：江苏省普通高校研究生科研创新计划项目(CXLX12_0628)
【分类号】：U491.61;U463.6
【正文快照】： 安全带作为乘员约束系统最有效的保护装置,自发明以来已经挽救了无数人的生命[1].为了提高安全带佩戴率,世界各国都强制安装了安全带佩戴提示系统(SRS).据A-NCAP公布的最新数据显示,澳大利亚80%以上的车型都配备了安全带佩戴提示系统[2].我国C-NCAP管理规则[3]中明确将安全带

【参考文献】