基于深度学习的水果缺陷实时检测方法  

Real time detection method of fruit defects based on deep learning

作　　者：周胜安[1] 黄耿生[1] 张译匀[1] 高东发 ZHOU Sheng-an;HUANG Geng-sheng;ZHANG Yi-yun;GAO Dong-fa(Guangdong Vocational Institute of Public Administration Electronic Information System,Guangzhou,Guangdong 510800,China;School of Computer Science,Guangdong University of Foreign Studies,Guangzhou,Guangdong 510665,China)

机构地区：[1]广东行政职业学院人工智能学院,广东广州510800 [2]广东外贸外语大学计算机科学学院,广东广州510665

出　　处：《食品与机械》2021年第11期123-129,共7页Food and Machinery

基　　金：广东省普通高校特色创新类项目(编号:2019GKTSCX053);广东省普通高校青年创新人才类项目(编号:2018GkQNCX125);中国高校产学研创新基金“新一代信息技术创新项目”(编号:2020ITA03013)。

摘　　要：目的:对CenterNet方法进行优化改进。方法:使用MobileNetV3的轻量化卷积神经网络替代CenterNet原有的骨干网络,加快检测速度,对MobileNetV3模块进行改进,增强模型对水果中小缺陷块的检测能力,并对CenterNet的预检测阶段进行优化,提高其检测准确度。结果:试验方法对显著缺陷如直径>4 mm的苹果识别率高达99.7%,检测速度为113帧/s,模型体积为1.31 MB。结论:与CenterNet_ResNet18和CenterNet_Shuffler模型相比,MO-CenterNet模型检测水果缺陷在训练时间、检测速度和准确率方面均衡性更好。Objective:This study aimed to optimize and improve the CenterNet method for detecting fruit defects.Methods:the lightweight convolutional neural network of MobileNetV3 was used to replace the original backbone network of CenterNet accelerate the detection speed,improve the module of MobileNetV3,enhance the detection ability of the model for small and medium-sized defective blocks of fruit,and optimize the pre detection stage of CenterNet to increase its detection accuracy.Results:The recognition rate of significant defects such as apples with diameter>4 mm was 99.7%,and the detection speed was 113 FPS,with the model volume of 1.31 MB.Conclusion:Compared with models CenterNet_Resnet18 and CenterNet_Shuffler,model MO-CenterNet has better balance in training time,detection speed and accuracy.

关 键 词：水果缺陷 实时检测 深度学习 卷积神经网络 轻量级 

分 类 号：TS255.7[轻工技术与工程—农产品加工及贮藏工程] TP18[轻工技术与工程—食品科学与工程] TP391.41[自动化与计算机技术—控制理论与控制工程]