摘 要:近年来,我国老年人人口数量占全国人口数量的比例持续上升,视障人群的占比也随之持续增加。针对这一问题,文章设计一款面向视障人群的视听辅助装置。该装置以 Mega2560 为核心控制器,与视觉模块、超声波模块、语音合成模块等共同实现目标检测、测距、语音播报、屏幕显示等功能。选用卷积神经网络进行目标检测与识别,提高了目标识别的精度。所设计的装置提高了视障人群的便利性与安全感,具有很高的应用价值。
关键词:视障人群;目标检测;卷积神经网络;语音播报
DOI:10.19850/j.cnki.2096-4706.2022.10.012
基金项目: 中原科技创新领军人才(214200510013); 国家自然科学基金项目(62171318); 河南省高校重点科研项目(21A510016,21A520052);留学人员科研资助和创业启动项目(HRSS2021[36]);校内重大项目成果培育计划(K2020ZDPY02)
中图分类号:TP391.4 文献标识码:A 文章编号:2096-4706(2022)10-0050-04
Design of Audio-visual Aid for Visually Impaired People
LI Gaoyang1, SONG Xiaowei 2, WEN Xiujing1, NIU Xu1
(1.School of Electronic and Information, Zhongyuan University of Technology, Zhengzhou 450007, China;2.Kaifeng University, Kaifeng 475004, China)
Abstract: In recent years, the proportion of the elderly population in the national population has continued to rise, and the proportion of the visually impaired population has also continued to increase. To solve this problem, this paper designs an audio-visual aid for visually impaired people. The device takes Mega2560 as the core controller and realizes the functions of target detection, ranging, voice broadcasting, screen display and so on together with vision module, ultrasonic module and voice synthesis module. The convolutional neural network is selected for target detection and recognition, which improves the accuracy of target recognition. The designed device improves the convenience and sense of security of visually impaired people, and has high application value.
Keywords: visually impaired people; target detection; convolutional neural network; voice broadcast
参考文献:
[1] 诗婧.盲人出行雷达语音辅助器设计研究 [J].科技与创新,2021(8):49-51.
[2] 刘金 . 盲人行走辅助技术中的障碍物检测算法研究 [D].成都:西南交通大学,2014.
[3] KAMMOUN S, PARSEIHIAN G, GUTIERREZ O,et al. Navigation and space perception assistance for the visually impaired: The NAVIG project [J].Irbm,2012,33(2):182-189.
[4] 方仁杰,朱维兵 . 基于 GPS 定位与超声波导盲拐杖的设计[J]. 计算机测量与控制,2011,19(5):1154-1157.
[5] 汪淑贤,李明,王旬 . 基于 Arduino Mega 2560 的水下探测仪设计与实现 [J]. 信息技术与网络安全,2020,39(4):48-52.
[6] YU X Q,SI Y J,LI L L. Pedestrian detection based on improved Faster RCNN algorithm [C]//2019 IEEE/CIC International Conference on Communications in China (ICCC).Changchun:IEEE,2019:2377-8644.
作者简介:李杲阳(1996—),男,汉族,河南南阳人,硕士研究生在读,主要研究方向:计算机视觉、深度学习;宋晓炜(1978—),男,汉族,山西大同人,教授,博士,主要研究方向:图像处理、计算机视觉;文秀静(1997—),女,汉族,河南新乡人,硕士研究生在读,主要研究方向:计算机视觉、深度学习;牛旭(1996—),男,汉族,河南驻马店人,硕士研究生在读,主要研究方向:计算机视觉、深度学习。