The invention relates to the technical field of driving safety, in particular to an automobile driver distracted driving early warning method based on a neural network, and the method comprises the steps: S1, building a distracted driving recognition model in a vehicle machine; s2, adopting a Grad-CAM algorithm to explain the distraction driving identification model; s3, training and testing the distraction driving identification model by using the data set, and analyzing and evaluating an identification result; and S4, acquiring behavior images of the driver in real time. The distraction driving behavior recognition model is established through deep transfer learning, the recognition key point of the established distraction driving recognition model is visualized, the distraction driving behavior recognition problem is well solved under the condition that high computer resources are not consumed, distraction can be recognized, the distraction type can also be recognized, and the recognition efficiency is improved. A feasible way is provided for deployment on a driving assistance system, the mode of using a traditional physiological sensor is avoided, and a better use prospect is brought.
本发明涉及行驶安全技术领域,具体涉及一种基于神经网络的汽车驾驶员分心驾驶预警方法,预警步骤为:S1、在车机内建立出分心驾驶识别模型;S2、采用Grad‑CAM算法对分心驾驶识别模型进行解释;S3、使用数据集对分心驾驶识别模型进行训练与测试,对识别结果进行分析与评估;S4、实时采集驾驶员的行为图像。本发明通过深度迁移学习,建立了分心驾驶行为识别模型,对建立的分心驾驶识别模型识别重点进行可视化,在不花费高额计算机资源的情况下很好的解决了分心驾驶行为识别问题,不仅能识别分心,也能识别分心的种类,为在驾驶辅助系统上部署提供了可行之道,避免了使用传统生理传感器的方式,带来了更好的使用前景。
Automobile driver distracted driving early warning method based on neural network
一种基于神经网络的汽车驾驶员分心驾驶预警方法
2024-06-28
Patent
Electronic Resource
Chinese
IPC: | G06V / B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen |
Vehicle-mounted driver fatigue distracted driving early warning method
European Patent Office | 2022
|DISTRACTED DRIVER DETECTION, CLASSIFICATION, WARNING, AVOIDANCE SYSTEM
European Patent Office | 2018
|