Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature Fusion
To achieve accurate detection the content of multiple parts pork adulterated in mutton under the effect of mutton flavor essence and colorant by RGB images, the improved CBAM-Invert-ResNet50 network based on the attention mechanism and the inversion residual was used to detect the content of pork fr...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-09-01
|
Series: | Foods |
Subjects: | |
Online Access: | https://www.mdpi.com/2304-8158/12/19/3594 |
_version_ | 1797575865725353984 |
---|---|
author | Zongxiu Bai Rongguang Zhu Dongyu He Shichang Wang Zhongtao Huang |
author_facet | Zongxiu Bai Rongguang Zhu Dongyu He Shichang Wang Zhongtao Huang |
author_sort | Zongxiu Bai |
collection | DOAJ |
description | To achieve accurate detection the content of multiple parts pork adulterated in mutton under the effect of mutton flavor essence and colorant by RGB images, the improved CBAM-Invert-ResNet50 network based on the attention mechanism and the inversion residual was used to detect the content of pork from the back, front leg, and hind leg in adulterated mutton. The deep features of different parts extracted by the CBAM-Invert-ResNet50 were fused by feature, stitched, and combined with transfer learning, and the content of pork from mixed parts in adulterated mutton was detected. The results showed that the <i>R</i><sup>2</sup> of the CBAM-Invert-ResNet50 for the back, front leg, and hind leg datasets were 0.9373, 0.8876, and 0.9055, respectively, and the RMSE values were 0.0268 g·g<sup>−1</sup>, 0.0378 g·g<sup>−1</sup>, and 0.0316 g·g<sup>−1</sup>, respectively. The <i>R</i><sup>2</sup> and RMSE of the mixed dataset were 0.9264 and 0.0290 g·g<sup>−1</sup>, respectively. When the features of different parts were fused, the <i>R</i><sup>2</sup> and RMSE of the CBAM-Invert-ResNet50 for the mixed dataset were 0.9589 and 0.0220 g·g<sup>−1</sup>, respectively. Compared with the model built before feature fusion, the <i>R</i><sup>2</sup> of the mixed dataset increased by 0.0325, and the RMSE decreased by 0.0070 g·g<sup>−1</sup>. The above results indicated that the CBAM-Invert-ResNet50 model could effectively detect the content of pork from different parts in adulterated mutton as additives. Feature fusion combined with transfer learning can effectively improve the detection accuracy for the content of mixed parts of pork in adulterated mutton. The results of this study can provide technical support and a basis for maintaining the mutton market order and protecting mutton food safety supervision. |
first_indexed | 2024-03-10T21:44:20Z |
format | Article |
id | doaj.art-1759c214fb6d439e8597d4b0b97cf8e0 |
institution | Directory Open Access Journal |
issn | 2304-8158 |
language | English |
last_indexed | 2024-03-10T21:44:20Z |
publishDate | 2023-09-01 |
publisher | MDPI AG |
record_format | Article |
series | Foods |
spelling | doaj.art-1759c214fb6d439e8597d4b0b97cf8e02023-11-19T14:23:04ZengMDPI AGFoods2304-81582023-09-011219359410.3390/foods12193594Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature FusionZongxiu Bai0Rongguang Zhu1Dongyu He2Shichang Wang3Zhongtao Huang4College of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, ChinaCollege of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, ChinaCollege of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, ChinaCollege of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, ChinaCollege of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, ChinaTo achieve accurate detection the content of multiple parts pork adulterated in mutton under the effect of mutton flavor essence and colorant by RGB images, the improved CBAM-Invert-ResNet50 network based on the attention mechanism and the inversion residual was used to detect the content of pork from the back, front leg, and hind leg in adulterated mutton. The deep features of different parts extracted by the CBAM-Invert-ResNet50 were fused by feature, stitched, and combined with transfer learning, and the content of pork from mixed parts in adulterated mutton was detected. The results showed that the <i>R</i><sup>2</sup> of the CBAM-Invert-ResNet50 for the back, front leg, and hind leg datasets were 0.9373, 0.8876, and 0.9055, respectively, and the RMSE values were 0.0268 g·g<sup>−1</sup>, 0.0378 g·g<sup>−1</sup>, and 0.0316 g·g<sup>−1</sup>, respectively. The <i>R</i><sup>2</sup> and RMSE of the mixed dataset were 0.9264 and 0.0290 g·g<sup>−1</sup>, respectively. When the features of different parts were fused, the <i>R</i><sup>2</sup> and RMSE of the CBAM-Invert-ResNet50 for the mixed dataset were 0.9589 and 0.0220 g·g<sup>−1</sup>, respectively. Compared with the model built before feature fusion, the <i>R</i><sup>2</sup> of the mixed dataset increased by 0.0325, and the RMSE decreased by 0.0070 g·g<sup>−1</sup>. The above results indicated that the CBAM-Invert-ResNet50 model could effectively detect the content of pork from different parts in adulterated mutton as additives. Feature fusion combined with transfer learning can effectively improve the detection accuracy for the content of mixed parts of pork in adulterated mutton. The results of this study can provide technical support and a basis for maintaining the mutton market order and protecting mutton food safety supervision.https://www.mdpi.com/2304-8158/12/19/3594adulterated muttonquantitative detectionsmart phonedeep learningCBAM-Invert-ResNet |
spellingShingle | Zongxiu Bai Rongguang Zhu Dongyu He Shichang Wang Zhongtao Huang Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature Fusion Foods adulterated mutton quantitative detection smart phone deep learning CBAM-Invert-ResNet |
title | Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature Fusion |
title_full | Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature Fusion |
title_fullStr | Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature Fusion |
title_full_unstemmed | Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature Fusion |
title_short | Adulteration Detection of Pork in Mutton Using Smart Phone with the CBAM-Invert-ResNet and Multiple Parts Feature Fusion |
title_sort | adulteration detection of pork in mutton using smart phone with the cbam invert resnet and multiple parts feature fusion |
topic | adulterated mutton quantitative detection smart phone deep learning CBAM-Invert-ResNet |
url | https://www.mdpi.com/2304-8158/12/19/3594 |
work_keys_str_mv | AT zongxiubai adulterationdetectionofporkinmuttonusingsmartphonewiththecbaminvertresnetandmultiplepartsfeaturefusion AT rongguangzhu adulterationdetectionofporkinmuttonusingsmartphonewiththecbaminvertresnetandmultiplepartsfeaturefusion AT dongyuhe adulterationdetectionofporkinmuttonusingsmartphonewiththecbaminvertresnetandmultiplepartsfeaturefusion AT shichangwang adulterationdetectionofporkinmuttonusingsmartphonewiththecbaminvertresnetandmultiplepartsfeaturefusion AT zhongtaohuang adulterationdetectionofporkinmuttonusingsmartphonewiththecbaminvertresnetandmultiplepartsfeaturefusion |