摘要
本文设计和实现了一款基于Android的垃圾分类助手APP,主要有三大功能,包括文本输入查询、语音识别查询和视频实时识别查询。其中,文本输入查询功能根据用户输入的垃圾名称来查询垃圾类别;语音识别查询功能借助讯飞语音流式识别功能,将音频信息转化为文本,再将查询结果反馈给用户;视频实时识别功能则直接获取智能手机捕获到的视频信息,借助谷歌公司提供的TensorFlow Lite框架对视频中的物品进行识别,再根据识别结果查询垃圾类别。
This paper designs and implements an Android based garbage sorting assistant APP,which has three main functions,including text input query,voice recognition query and real-time video recognition query.Among them,the text input query function queries the garbage category according to the spam name entered by the user;the voice recognition query function converts the audio information into text with the help of iFLYTEK voice flow recognition function,and then feeds back the query results to the users;the video real-time recognition function directly obtains the video information captured by the smart phone,with the help of TensorFlow provided by Google The Lite framework identifies the objects in the video,and then queries the garbage categories according to the identification results.
作者
高文超
王子聪
刘昊天
Gao Wenchao;Wang Zicong;Liu Haotian(China University of Mining&Technology,Beijing,Beijing 100083,China)
出处
《信息与电脑》
2020年第19期58-61,共4页
Information & Computer
基金
中央高校基本科研业务费专项资金资助(项目编号:2020YQJD15)
中国矿业大学(北京)本科教育教学改革与研究项目建设(项目编号:J200513)。
