The invention provides a vehicle-mounted voice wake-up method and system based on facial action perception. The method comprises the following steps: acquiring a facial image of a user in a vehicle; the face image is preprocessed; performing facial action recognition on the preprocessed facial image to obtain a preset facial action set; sending a wake-up instruction of a vehicle-mounted voice system according to the specified facial personalized action in the preset facial action set; and activating the vehicle-mounted voice system according to the wake-up instruction. According to the invention, the problem that the current mainstream vehicle-mounted voice system wake-up mode is easily interfered by external noise, is inconvenient to operate and has driving potential safety hazards, so that the vehicle-mounted voice wake-up is not intelligent enough can be solved.
本发明提出一种基于面部动作感知的车载语音唤醒方法及系统,该方法包括:采集车内用户的面部图像;对所述面部图像进行预处理;将所述预处理后的面部图像进行面部动作识别,获得预设面部动作集;根据所述预设面部动作集中指定的面部个性动作发送车载语音系统的唤醒指令;根据所述唤醒指令激活车载语音系统。本发明能够解决目前主流的车载语音系统唤醒方式容易受外界噪音干扰、操作不便以及有驾驶安全隐患,使得车载语音唤醒不够智能的问题。
Vehicle-mounted voice wake-up method and system based on facial action perception
一种基于面部动作感知的车载语音唤醒方法及系统
2024-06-18
Patent
Electronic Resource
Chinese
Vehicle-mounted interaction system based on emotion perception and voice interaction
European Patent Office | 2022
|Vehicle-mounted wake-up circuit and vehicle-mounted machine wake-up system
European Patent Office | 2025
|