论文部分内容阅读
建立交通事故应急救援系统具有很强的现实意义,关键技术是对交通事故进行自动检测。为了保证检测的实时性与准确性,提出一种基于声信号处理的方法,采集车辆周围的声音并进行预处理,使用Harr小波变换提取声信号的频域特征,采用单类支持向量机进行异常点检测,实现了分类判别。按上述方法对交通事故发生时的碰撞信号与正常行驶时的非碰撞信号做了分析,准确的识别出交通事故。仿真实验结果表明与常用的线性判别分析方法相比准确率有了显著提高,而且计算复杂度低,易于在DSP系统上实现,算法的判别性能达到了实用化的程度。
The establishment of emergency rescue system for traffic accidents has strong practical significance. The key technology is automatic detection of traffic accidents. In order to ensure the real-time performance and accuracy of detection, a method based on acoustic signal processing is proposed. The sound around the vehicle is collected and preprocessed. Harr wavelet transform is used to extract the frequency domain features of acoustic signals. Single-class support vector machines Point detection, to achieve a classification of discrimination. According to the above method, the collision signal at the time of traffic accident and the non-collision signal at normal driving time are analyzed, and the traffic accident is accurately identified. Simulation results show that compared with the commonly used linear discriminant analysis, the accuracy is significantly improved, and the computational complexity is low and easy to implement on the DSP system. The discriminative performance of the algorithm reaches a practical level.