[1]王 君,梁文威*,麦淳铭,等.智能图像处理与应用设计实例[J].计算机技术与发展,2021,31(12):217-220.[doi:10. 3969 / j. issn. 1673-629X. 2021. 12. 036]
　WANG Jun,LIANG Wen-wei*,MAI Chun-ming,et al.Design of Intelligent Image Processing and Application[J].,2021,31(12):217-220.[doi:10. 3969 / j. issn. 1673-629X. 2021. 12. 036]
点击复制

智能图像处理与应用设计实例()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
31
期数:
2021年12期
页码:
217-220
栏目:
应用前沿与综合
出版日期:
2021-12-10

文章信息/Info

Title:
Design of Intelligent Image Processing and Application
文章编号:
1673-629X(2021)12-0217-04
作者:
王 君1; 2 ; 梁文威1* ; 麦淳铭1 ; 宋泽生3 ; 梁薇薇4 ; 蒲 磊1
1. 中山大学南方学院,广东 广州 510970;
2. 广州恒通智联科技有限公司,广东 广州 510630;
3. 南昌大学,江西 南昌 330027;
4. 重庆邮电大学,重庆 400065
Author(s):
WANG Jun1; 2 ; LIANG Wen-wei1* ; MAI Chun-ming1 ; SONG Ze-sheng3 ; LIANG Wei-wei4 ; PU Lei1
1. Nanfang College of Sun Yat-sen University,Guangzhou 510970,China;
2. Guangzhou Hengtong Zhilian Technology Limited Company,Guangzhou 510630,China;
3. Nanchang University,Nanchang 330027,China;
4. Chongqing University of Posts and Telecommunications,Chongqing 400065,China
关键词:
智能图像处理; Tensorflow 算法; 卷积神经网络分类识别; 手机识别; 智能化回收利用; 绿色环保
Keywords:
intelligent image processing; Tensorflow algorithm; convolutional neural network classification and recognition; mobile phonerecognition; energy recycling; green environmental protection
分类号:
TP305
DOI:
10. 3969 / j. issn. 1673-629X. 2021. 12. 036
文献标志码:
A
摘要:
智能图像处理技术紧跟科技前沿,其算法被广泛应用于实际生产中,带来了很多的便利和经济价值,尤其是计算机视觉图像的识别技术。 针对废旧手机堆积如山,造成大量资源浪费、污染环境的问题,以对手机进行分类回收为例,文中利用机器视觉对手机完成快速分类,以便于提取贵重金属、稀有金属,充分利用资源,节省人力成本。 该方法基于Tensorflow 算法,对手机进行图像训练分类,形成模板库,让待识别手机与模板库进行匹配,利用卷积神经网络分类识别方法进行识别,最终得以成功应用。 以 Linux 系统的树莓派 3B+为主控制器,32 位 ARMCotex-M3 内核单片机为辅助控制器,实现舵机的转动,自动完成手机识别与分类回收。 该方法能快速有效地识别和分类手机,节省分类手机的时间,且可以流水作业,便于贵重金属和稀有金属的智能化回收利用。
Abstract:
Intelligent image processing technology keeps up with the frontier of science and technology,and its algorithm is widely used in production,which brings a lot of convenience and economic value,especially the image recognition technology based on computer vision.Aiming at the problem of waste mobile phones piling up,resulting in a large amount of resource waste and environmental pollution,taking the classification and recycling of mobile phones as an example,we complete rapid classification of mobile phones according to machine vision,in order to extract precious metals and rare metals,make full use of resources and save labor costs. Based on Tensorflow,the proposed method carries out image training and classification on mobile phones to form a template library,and matches the mobile phones to be identified with the template library,and uses the convolutional neural network classification and recognition method to identify.Finally,it is successfully applied. The Raspberry Pi 3B+ of Linux system is used as the main controller,and the 32-bit ARMCotex-M3 kernel MCU is used as the auxiliary controller to realize the rotation of the steering gear and automatically complete the identification and classification of mobile phones. The proposed method can quickly and effectively identify and classify mobile phones,save the time of classifying mobile phones,and can be used in production line,which is convenient for intelligent recycling of precious metals and rare metals.
更新日期/Last Update: 2021-12-09