Abstract
Objective: In order to solve the problem of edible fungus species identification, an EfficientNet edible fungus image classification model based on convolution neural network is proposed. Methods: Firstly, the edible fungus images were collected and the datasets were made according to different equipment and shooting environment, and then the model performance was improved through model training skills and network skills. A YWeight weight attenuation method was proposed to control the effective learning rate, and the generalization performance of the model was affected by controlling the cross-boundary. Results: This method makes EfficientNet-B0 obtain 79.82% (+0.85%) top-1 accuracy on the self built dataset YMushroom, and only 78.97% in the default training process. On the public dataset fungus, the accuracy of EfficientNet-B0 was 87.62% (+0.78%) and the original training accuracy was 86.84%. Conclusion: Experiments show that by adjusting the super parameters, the model finds a near optimal solution, and improves the performance of the edible fungus image classification model through weight attenuation, which provides a basis for the automatic management of edible fungus planting base in the future.
Publication Date
12-15-2022
First Page
117
Last Page
124
DOI
10.13652/j.spjx.1003.5788.2022.90082
Recommended Citation
Zhi-xin, YAO; Tai-hong, ZHANG; and Yun-jie, ZHAO
(2022)
"EfficientNet edible fungus image recognition based on improved weight decay,"
Food and Machinery: Vol. 38:
Iss.
11, Article 20.
DOI: 10.13652/j.spjx.1003.5788.2022.90082
Available at:
https://www.ifoodmm.cn/journal/vol38/iss11/20
References
[1] 鲍大鹏.食用菌科学研究为食用菌产业发展提供越来越重要的科学支撑[J].菌物学报,2021,40(2):3 061-3 063.BAO D P.Scientific research on edible fungi provides more and more important scientific support for the development of edible fungi industry[J].Mycosystema,2021,40(2):3 061-3 063.
[2] 薛雨.基于机器视觉技术的食用菌成长过程监控系统研究[D].郑州:华北水利水电大学,2018:55.XUE Y.Research on monitoring system of edible fungi growth process based on machine vision technology[D].Zhengzhou:North China University of Water Resources and Electric Power,2018:55.
[3] 林楠,王娜,李卓识,等.基于机器视觉的野生食用菌特征提取识别研究[J].中国农机化学报,2020,41(5):111-119.LIN N,WANG N,LI Z S,et al.Research on feature extraction and recognition of wild edible fungi based on machine vision[J].Journal of Chinese Agricultural Mechanization,2020,41(5):111-119.
[4] FILIPPOVA N,BULYONKOVA T,BANAEV E V,et al.Fungal records database of khanty-mansi autonomous okrug-yugra[J].BIO Web of Conferences,2018,11:15-20.
[5] 智研咨询集团.2020—2026年中国食用菌行业市场行情监测及发展前景展望报告[EB/OL].(2020-02-14)[2021-10-17].https://www.chyxx.com/industry/202002/833926.html.Zhiyan Cousulting Group.Report on market monitoring and development prospect of China’s edible fungi industry from 2020 to 2026[EB/OL].(2020-02-14)[2021-10-17].https://www.chyxx.com/industry/202002/833926.html.
[6] MARMANIS D,DATCU M,ESCH T,et al.Deep learning earth observation classification using ImageNet pretrained networks[J].IEEE Geoscience & Remote Sensing Letters,2016,13(1):105-109.
[7] SON H,FONG Y.Fast grid search and bootstrap-based inference for continuous two-phase polynomial regression models[J].Environmetrics,2021,32(3):e2664.1-e2664.16.
[8] 吴慧华,苏寒松,刘高华,等.基于余弦距离损失函数的人脸表情识别算法[J].激光与光电子学进展,2019,56(24):196-202.WU H H,SU H S,LIU G H,et al.Facial expression recognition algorithm based on cosine distance loss function[J].Laser & Optoelectronics Progress,2019,56(24):196-202.
[9] 章东平,陈思瑶,李建超,等.基于改进型加性余弦间隔损失函数的深度学习人脸识别[J].传感技术学报,2019,32(12):1 830-1 835.ZHANG D P,CHEN S Y,LI J C,et al.Deep learning face recognition based on improved additive cosine interval loss function[J].Chinese Journal of Sensors and Actuators,2019,32(12):1 830-1 835.
[10] NGO N,PORTER G A,CHERNG H Q,et al.Development of a color object classification and measurement system using machine vision[J].Sensors and Materials,2019,31(12):4 135-4 154.
[11] LI W,LIU K.Confidence-aware object detection based on MobileNetv2 for autonomous driving[J].Sensors,2021,21(7):2 380.
[12] MALEKABADI A J,KHOJASTEHPOUR M,EMADI B,et al.Development of a machine vision system for determination of mechanical properties of onions[J].Computers & Electronics in Agriculture,2017,141:131-139.
[13] 郭丽丽,丁世飞.深度学习研究进展[J].计算机科学,2015,42(5):28-33.GUO L L,DING S F.Research progress of deep learning[J].Computer Science,2015,42(5):28-33.
[14] ATILA M,UAR M,AKYOL K,et al.Plant leaf disease classification using EfficientNet deep learning model[J].Ecological Informatics,2021,61:101182.
[15] DUONG L T,NGUYEN P T,DI SIPIO C,et al.Automated fruit recognition using EfficientNet and MixNet[J].Computers and Electronics in Agriculture,2020,171:105326.
[16] 李文宝,曹成茂,张金炎,等.基于深度学习的山核桃破壳物料分类识别[J].食品与机械,2021,37(9):133-138,185.LI W B,CAO C M,ZHANG J Y,et al.Classification and recognition of pecan shell breaking materials based on deep learning[J].Food & Machinery,2021,37(9):133-138,185.
[17] MA H,CHEN M,ZHANG J W.Study on the fruit grading recognition system based on machine vision[J].Advance Journal of Food Science and Technology,2015,8(11):777-780.
[18] 赵世达,王树才,李振强,等.基于U型卷积神经网络的羊肋排图像分割[J].食品与机械,2020,36(9):116-121,154.ZHAO S D,WANG S C,LI Z Q,et al.Segmentation of sheep rib image based on U-shaped convolution neural network[J].Food & Machinery,2020,36(9):116-121,154.
[19] 张璐,李卓识,李玉.鹅膏属真菌形态特征的主成分与聚类分析[J].菌物学报,2018,37(5):559-564.ZHANG L,LI Z S,LI Y.Principal component and cluster analysis of morphological characteristics of Amanita fungi[J].Mycosystema,2018,37(5):559-564.
[20] 王海燕,张渺,刘虎林,等.基于改进的ResNet网络的中餐图像识别方法[J].陕西科技大学学报,2022,40(1):154-160.WANG H Y,ZHANG M,LIU H L,et al.Chinese food image recognition method based on improved ResNet[J].Journal of Shaanxi University of Science & Technology,2022,40(1):154-160.