The invention provides a walking assistance system. The signal lamp recognition device is based on a result of a recognition operation of the signal lamp first recognition unit (a recognition operation of the signal lamp using a learned model based on pre-annotated data for an image acquired by the camera) and a recognition operation of the signal lamp second recognition unit (a recognition operation of the signal lamp using a learned model based on the pre-annotated data for an image acquired by the camera). And determines whether or not a signal lamp is recognized in the image based on the signal lamp feature amount. As a result, it is possible to sufficiently obtain the recognition accuracy of the traffic light, and to appropriately give an instruction to a pedestrian in accordance with the state of the traffic light.
本发明提供一种步行辅助系统。分别基于信号灯第一识别部的识别动作(对于由摄像机取得的图像,使用基于预先注释的数据的已学习模型的信号灯的识别动作)的结果、以及信号灯第二识别部的识别动作(对于由摄像机取得的图像,基于信号灯的特征量的该信号灯的识别动作)的结果,判定在图像中是否识别出信号灯。由此,能够充分地得到信号灯的识别精度,从而能够适当地进行与信号灯的状态相应的对步行者的指示。
Walking assistance system
步行辅助系统
2022-11-15
Patent
Electronic Resource
Chinese