Detection System Study of Defective Egg on Mobile Devices Based on Deep Learning

被引：0

作者：

Fan W. ^{[1
]}

Hu J. ^{[1
]}

Wang Q. ^{[1
,2
]}

Tang W. ^{[1
]}

机构：

[1] College of Engineering, Huazhong Agricultural University, Wuhan

[2] Key Laboratory of Agricultural Equipment in Mid-lower Yangtze River, Ministry of Agriculture and Rural Affairs, Wuhan

来源：

Nongye Jixie Xuebao/Transactions of the Chinese Society for Agricultural Machinery | 2023年 / 54卷 / 03期

关键词：

deep learning; defective egg; mobile devices; model optimization; non-destructive detection;

D O I：

10.6041/j.issn.1000-1298.2023.03.042

中图分类号：

学科分类号：

摘要：

Aiming at the problems of large diversity of defective eggs, as well as the strong subjectivity and poor real-time detection of artificial detection, and the potential risk of food safety for end-consumers, a non-destructive testing system based on deep learning for defective eggs on mobile device was proposed to realize real-time detection of cracked eggs and bloody eggs. An improved lightweight convolutional neural network MobileNetV2-CA model was firstly established. MobileNetV2 network was taken as the original framework, it was further optimized by embedding coordinate attention mechanism, adjusting width factor, transfer learning and other parameters. The PC detection was also performed for comparison. Results showed that the MobileNetV2 - CA model presented the validation accuracy of 93. 93%, the recall rate of 94. 73%, and the average detection time of 9.9 ms for a single egg, which was 3.60 percentage points higher, 4.30 percentage points higher, and 2.62 ms shorter than the original MobileNetV2 model, respectively. The parameter score of MobileNetV2-CA model was only 2.36 × 106, which was 31.59% lower than the original MobileNetV2 network model. In addition, the NCNN deep learning framework was used to train MobileNetV2-CA model, which was further applied to Android mobile terminal through format conversion. The verification of mobile terminal detection of NCNN deep learning training model was investigated and compared with TensorFlow Lite deep learning model. Results showed that the NCNN deep learning model had an average recognition accuracy of 92. 72%, an average detection time of 22.1 ms for a single egg, and the library file size of 2.7 MB, indicating its better performance than TensorFlow Lite and meeting the requirement of practical applications. The effectiveness of the proposed system based on deep learning was finally demonstrated. © 2023 Chinese Society of Agricultural Machinery. All rights reserved.

引用

页码：411 / 420

页数：9

共 27 条

[11]

WANG Qiaohua, WANG Caiyun, MA Meihu, Duck eggs' freshness detection based on machine vision technology [J], Journal of Chinese Institute of Food Science and Technology, 17, 8, pp. 268-274, (2017)

[12]

HOWARD A G, ZHU M, CHEN B, Et al., MobileNets: efficient convolutional neural networks for mobile vision applications [J], (2017)

[13]

SANDLER M, HOWARD A, ZHU M, Et al., MobileNetV2: inverted residuals and linear bottlenecks[C], Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 4510-4520, (2018)

[14]

SUHARJITO B, GNE A, JSP A., Oil palm fresh fruit bunch ripeness classification on mobile devices using deep learning approaches[J], Computers and Electronics in Agriculture, 188, (2021)

[15]

HOU Q, ZHOU D, FENG J., Coordinate attention for efficient mobile network design, (2021)

[16]

HU Jie, SHEN Li, SUN Gang, Et al., Squeeze-and-excitation networks [J/OL], IEEE Transactions on Pattern Analysis and Machine Intelligence, (2019)

[17]

WOO S, PARK J, LEE J Y, Et al., CBAM: convolutional block attention module [C], European Conference on Computer Vision, (2018)

[18]

ZHOU Zhongxian, SONG Zhenzhen, FU Longsheng, Et al., Real-time kiwifruit detection in orchard using deep learning on Android smartphones for yield estimation[J], Computers and Electronics in Agriculture, 179, (2020)

[19]

LE T D, BERCEA G T, CHEN T, Et al., Compiling ONNX neural network models using MLIR, (2020)

[20]

GU Diandian, SHI Yining, LIU Huanzhe, Et al., Defect detection for deep learning frameworks based on meta operators [J], Chinese Journal of Computers, 45, 2, pp. 240-255, (2022)

← 1 2 3 →