基于人工智能技术的多信息融合可穿戴式导盲系统设计

Design of Multi-information Fusion Wearable Guide System Based on Artificial Intelligence

人获取外界信息主要通过视觉。失去视觉后,盲人在日常生活中存在诸多困难与危险。随着计算机软硬件技术的迅速发展,导盲设备层出不穷,使盲人群体的生活状况有所改善,切实享受到科技进步带来的便利。现有的导盲设备为盲人的生活提供一些帮助,但存在着价格高、体验差、功能单一等问题。基于以上原因,该文提出一种更人性化、大众化、智能化的基于人工智能技术的多信息融合可穿戴式导盲系统。系统选用带有Raspbian操作系统的树莓派4B开发板作为主控制器,协控制器选用STM32开发板,以OpenCV和C++作为系统的开发工具。系统中的物体识别部分采用OpenCV提供的图像处理接口实现,安全避障部分采用激光测距模块获取障碍物距离,语音识别部分采用语音识别模块对系统进行语音控制,可拨打紧急电话并发送带GPS定位的求救短信。项目采用多线程开发技术开发系统软件,使得系统的性能更加优越。

The main way for humans to obtain external information is through vision.With the loss of vision,the blind have many difficulties and dangers in their daily lives.With the rapid development of computer software and hardware technology,a variety of guide devices for the blind have emerged,improving the living conditions of the blind population and allowing them to truly enjoy the convenience brought by technological progress.The existing guide devices have provided some help to the blind in their lives.However,existing guide devices for the blind have problems such as high prices,poor user experience,and single function.For these reasons,this paper proposes a more humane,popular and intelligent wearable guide system based on artificial intelligence technology with multi-information fusion.The system uses Raspberry Pi 4B development board with Raspbian OS as the main controller,STM32 development board as the co-controller,OpenCV and C++as the development tools of the system.The object recognition part of the system is implemented using the image processing interface provided by OpenCV,the safety obstacle avoidance part uses a laser ranging module to obtain the distance to obstacles,and the speech recognition part uses a speech recognition module for voice control of the system,which can make emergency calls and send rescue messages with GPS positioning.The project adopts multi-threaded development technology to develop the system software,which makes the system's performance more superior.