An Intelligent Method for Pork Freshness Identification Based on EfficientNet Model
A method for measuring pork freshness based on images and the EfficientNet framework was established. A total of 2 500 images of pork with different freshness were collected as original dataset and processed by image enhancement to construct a new dataset of 60 000 images. First, EfficientNet was t...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
China Food Publishing Company
2023-12-01
|
| Series: | Shipin Kexue |
| Subjects: | |
| Online Access: | https://www.spkx.net.cn/fileup/1002-6630/PDF/2023-44-24-045.pdf |
| _version_ | 1797365335921262592 |
|---|---|
| author | LIU Chao, ZHANG Jiayu, QI Chao, HUANG Jichao, CHEN Kunjie |
| author_facet | LIU Chao, ZHANG Jiayu, QI Chao, HUANG Jichao, CHEN Kunjie |
| author_sort | LIU Chao, ZHANG Jiayu, QI Chao, HUANG Jichao, CHEN Kunjie |
| collection | DOAJ |
| description | A method for measuring pork freshness based on images and the EfficientNet framework was established. A total of 2 500 images of pork with different freshness were collected as original dataset and processed by image enhancement to construct a new dataset of 60 000 images. First, EfficientNet was trained with the CIFAR-10 dataset to determine the basic structure and initial weights of the model. Then, the model was trained and improved using the constructed dataset to make the model suitable for five classification problems. Finally, the established model was tested, verified, and compared with the current mainstream convolutional neural network (CNN) models of Alexnet, VGG16 and ResNet50. The results showed that the average correct recognition rate of the EfficientNet model was as high as 98.62%, which was significantly better than that of the Alexnet, VGG16 and ResNet50 models. The correct recognition rate of the EfficientNetB2 model was 99.22%, and the training time was only 157 min. The comprehensive performance of the EfficientNetB2 model was the best, making it the most suitable method for pork freshness identification. In order to improve its generalization ability, the optimizer algorithm of the EfficientNetB2 model was improved, and the performances of stochastic gradient descent (SGD), adaptive moment estimation (Adam), root mean square propagation (RMSProp) and rectified adaptive moment estimation (RAdam) were compared. The results showed that the RAdam optimizer failed to further improve the accuracy of the model but instead helped to improve its generalization capability, which will of practical significance for engineering applications. |
| first_indexed | 2024-03-08T16:47:40Z |
| format | Article |
| id | doaj.art-08a2b190b9164b9996f7d89656eb3aff |
| institution | Directory Open Access Journal |
| issn | 1002-6630 |
| language | English |
| last_indexed | 2024-03-08T16:47:40Z |
| publishDate | 2023-12-01 |
| publisher | China Food Publishing Company |
| record_format | Article |
| series | Shipin Kexue |
| spelling | doaj.art-08a2b190b9164b9996f7d89656eb3aff2024-01-05T07:16:27ZengChina Food Publishing CompanyShipin Kexue1002-66302023-12-01442436937710.7506/spkx1002-6630-20221218-182An Intelligent Method for Pork Freshness Identification Based on EfficientNet ModelLIU Chao, ZHANG Jiayu, QI Chao, HUANG Jichao, CHEN Kunjie0(1. College of Engineering, Nanjing Agricultural University, Nanjing 210031, China; 2. College of Intelligent Manufacturing, Taizhou Institute of Science and Technology, Nanjing University of Science and Technology, Taizhou 225300, China)A method for measuring pork freshness based on images and the EfficientNet framework was established. A total of 2 500 images of pork with different freshness were collected as original dataset and processed by image enhancement to construct a new dataset of 60 000 images. First, EfficientNet was trained with the CIFAR-10 dataset to determine the basic structure and initial weights of the model. Then, the model was trained and improved using the constructed dataset to make the model suitable for five classification problems. Finally, the established model was tested, verified, and compared with the current mainstream convolutional neural network (CNN) models of Alexnet, VGG16 and ResNet50. The results showed that the average correct recognition rate of the EfficientNet model was as high as 98.62%, which was significantly better than that of the Alexnet, VGG16 and ResNet50 models. The correct recognition rate of the EfficientNetB2 model was 99.22%, and the training time was only 157 min. The comprehensive performance of the EfficientNetB2 model was the best, making it the most suitable method for pork freshness identification. In order to improve its generalization ability, the optimizer algorithm of the EfficientNetB2 model was improved, and the performances of stochastic gradient descent (SGD), adaptive moment estimation (Adam), root mean square propagation (RMSProp) and rectified adaptive moment estimation (RAdam) were compared. The results showed that the RAdam optimizer failed to further improve the accuracy of the model but instead helped to improve its generalization capability, which will of practical significance for engineering applications.https://www.spkx.net.cn/fileup/1002-6630/PDF/2023-44-24-045.pdfpork freshness; non-destructive inspection; deep learning; efficientnet |
| spellingShingle | LIU Chao, ZHANG Jiayu, QI Chao, HUANG Jichao, CHEN Kunjie An Intelligent Method for Pork Freshness Identification Based on EfficientNet Model Shipin Kexue pork freshness; non-destructive inspection; deep learning; efficientnet |
| title | An Intelligent Method for Pork Freshness Identification Based on EfficientNet Model |
| title_full | An Intelligent Method for Pork Freshness Identification Based on EfficientNet Model |
| title_fullStr | An Intelligent Method for Pork Freshness Identification Based on EfficientNet Model |
| title_full_unstemmed | An Intelligent Method for Pork Freshness Identification Based on EfficientNet Model |
| title_short | An Intelligent Method for Pork Freshness Identification Based on EfficientNet Model |
| title_sort | intelligent method for pork freshness identification based on efficientnet model |
| topic | pork freshness; non-destructive inspection; deep learning; efficientnet |
| url | https://www.spkx.net.cn/fileup/1002-6630/PDF/2023-44-24-045.pdf |
| work_keys_str_mv | AT liuchaozhangjiayuqichaohuangjichaochenkunjie anintelligentmethodforporkfreshnessidentificationbasedonefficientnetmodel AT liuchaozhangjiayuqichaohuangjichaochenkunjie intelligentmethodforporkfreshnessidentificationbasedonefficientnetmodel |