| 李开放,刘忠涛,柏兴涛,张冰战.基于k-means聚类和神经网络的超速行为识别研究[J].安徽建筑大学学报,2022,30(6):83-88 |
| 基于k-means聚类和神经网络的超速行为识别研究 |
| Research on Speeding Behavior Recognition Based on K-means Clustering and Neural Network |
| |
| DOI： |
| 中文关键词: k-means算法 高斯混合聚类算法 神经网络 超速行为 |
| 英文关键词: k-means algorithm Gaussian mixture clustering algorithm neural network overspeed behavior |
| 基金项目:国家新能源汽车重点研发计划项目（2017YFB0103204） |
|
| 摘要点击次数: 4522 |
| 全文下载次数: 0 |
| 中文摘要: |
| 交通事故的发生对人们的生命财产安全造成了威胁，而超速驾驶是交通事故发生的一个重要因素。因此，如何准确识别超速行为至关重要。本文提出了一种基于工况识别的超速驾驶行为识别方法，首先利用主成分分析法对数据进行降维，利用k-means算法和高斯混合聚类算法对降维结果进行二次聚类，根据聚类结果训练BP神经网络，用训练好的模型对工况进行实时识别，进而得到不同工况的速度阈值用于超速行为识别。研究结果表明行驶工况的平均识别正确率达95%，将该方法应用于超速行为的识别，可使识别更加准确、科学。 |
| 英文摘要: |
| The occurrence of traffic accidents poses a threat to the safety of peoples’s lives and property, and speeding is a major factor in traffic accidents. Therefore, how to accurately identify speeding behavior is crucial. This paper proposes a speeding behavior recognition method based on working condition recognition. First, the principal component analysis method is used to reduce the dimensionality of the data, and the k-means algorithm and the Gaussian mixture clustering algorithm are used to perform secondary clustering of the dimensionality reduction results, BP neural network is trained according to the clustering results, and the trained model is used to recognize the operating conditions in real time, and then the speed thresholds of different working conditions are obtained for speeding behavior recognition. The research results show that the recognition accuracy of driving conditions reaches 95% on average, the application of this method to the recognition of speeding behaviour can make the recognition more accurate and scientific. |
|
查看全文
查看/发表评论 下载PDF阅读器 |
| 关闭 |
