摘 要:针对无线投影系统市场需求量较大这一情况,文章设计一款基于深度学习的手势识别系统,采用 ARM-A9 开发板进行开发,通过摄像头获取用户手部图像数据,运用 OpenCV 机器视觉库进行图像处理,再使用 tensorflow 深度神经网络对处理后的图像进行模型训练,从而更好地识别用户的手势姿态,结合无线投影技术实现手势识别技术的进一步应用。
关键词:人机交互;tensorflow 深度神经网络;OpenCV
DOI:10.19850/j.cnki.2096-4706.2022.011.027
基金项目:2020 年广东省“攀登计划”专项资金重点项目(pdjh2020a0859)
中图分类号:TP391.4 文献标识码:A 文章编号:2096-4706(2022)11-0106-04
Design and Implementation of Gesture Recognition System Based on Deep Learning
LI Xie, CAI Zhenxiong, ZHAN Wenjie, QIU Ziyi
(Software Engineering Institute of Guangzhoun (Guangzhou), Guangzhou 510980, China)
Abstract: In view of the large market demand for wireless projection system, this paper designs a gesture recognition system based on deep learning, which is developed with ARM-A9 development board. The user’s hand image data is obtained through the camera, and the OpenCV machine vision library is used for image processing. Then the processed image is model trained with tensorflow deep neural network, so as to better recognize the user’s gesture, combined with wireless projection technology, the further application of gesture recognition technology is realized.
Keywords: human computer interaction; tensorflow deep neural network; OpenCV
参考文献:
[1] 刘培军,马明栋,王得玉 . 基于 OpenCV 图像处理系统的开发与实现 [J]. 计算机技术与发展,2019,29(3):127-131.
[2] 张朝柱,顾晓婷,张艺漫 . 基于深度卷积神经网络的手势动作识别 [J]. 无线电工程,2019,49(7):587-591.
[3] TRIESCH J,MALSBURG C V D.Classification of hand postures against complex backgrounds using elastic graph matching [J].Image and Vision Computing,2002,20(13):937-943.
[4] 赵庆平,陈得宝,姜恩华,等 . 一种改进权重的非局部均值图像去噪算法 [J]. 电子测量与仪器学报,2014,28(3):334-339.
[5] 黄金平,张正炳 . 卷积定理与傅立叶变换性质及其应用的关系探讨 [J]. 长江大学学报(自科版),2016,13(19):29-31+41.
[6] 洑涵妤 . 基于嵌入式 Linux 系统的图像处理研究 [J]. 信息与电脑(理论版),2010(4):66-67.
作者简介:李澥(1963—),男,汉族,广东梅县人,副教授,本科,研究方向:物联网应用技术。