Green Apple Detection Method Based on Multidimensional Feature Extraction Network Model and Transformer Module
To enhance the fast and accurate detection of pollution-free green apples for food safety, this paper uses the DETR network as a framework to propose a new method for pollution-free green apple detection based on a multidimensional feature extraction network and Transformer module. Firstly, an impro...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Journal of Food Protection |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0362028X24001819 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841553931485839360 |
|---|---|
| author | Wei Ji Kelong Zhai Bo Xu Jiawen Wu |
| author_facet | Wei Ji Kelong Zhai Bo Xu Jiawen Wu |
| author_sort | Wei Ji |
| collection | DOAJ |
| description | To enhance the fast and accurate detection of pollution-free green apples for food safety, this paper uses the DETR network as a framework to propose a new method for pollution-free green apple detection based on a multidimensional feature extraction network and Transformer module. Firstly, an improved DETR network main feature extraction module adopts the ResNet18 network and replaces some residual layers with deformable convolutions (DCNv2), enabling the model to better adapt to pollution-free fruit changes at different scales and angles, while eliminating the impact of microbial contamination on fruit testing; Subsequently, the extended spatial pyramid pooling model (DSPP) and multiscale residual aggregation module (FRAM) are integrated, which help reduce feature noise and minimize the loss of underlying features during the feature extraction process. The fusion of the two modules enhances the model’s ability to detect objects of different scales, thereby improving the accuracy of near-color fruit detection. At the same time, in order to solve the problems of slow convergence speed and large calculation amount of the basic network model, the convergence speed of the overall network model is improved by replacing the attention mechanism of Transformer. Experimental results show that compared with the original DETR model, the proposed algorithm has improved in AP, AP50, and AP75 indicators, especially in the AP50 indicator, which has the most obvious improvement reaching a detection accuracy of 97.12%. In the meantime, the trained network model is deployed on the picking robot. Compared with the original DETR network model, its average detection accuracy is as high as 96.58%, and the detection speed is increased by about 51%. Mixed sample detection tests were carried out before and after the model deployment, and the detection rate of the proposed method for nonpolluted fruits reached more than 0.95. enabling the picking robot to efficiently complete the task of picking green apples. The test results show that the algorithm proposed in this article exhibits great potential in the task of detecting pollution-free near-color fruits by the picking robot. It ensures pollution-free fruit picking and the application of AI in food safety. |
| format | Article |
| id | doaj-art-ef5d971f093f47a7865a89539648b016 |
| institution | Kabale University |
| issn | 0362-028X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Food Protection |
| spelling | doaj-art-ef5d971f093f47a7865a89539648b0162025-01-09T06:12:29ZengElsevierJournal of Food Protection0362-028X2025-01-01881100397Green Apple Detection Method Based on Multidimensional Feature Extraction Network Model and Transformer ModuleWei Ji0Kelong Zhai1Bo Xu2Jiawen Wu3The School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, ChinaThe School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, ChinaCorresponding author.; The School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, ChinaThe School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, ChinaTo enhance the fast and accurate detection of pollution-free green apples for food safety, this paper uses the DETR network as a framework to propose a new method for pollution-free green apple detection based on a multidimensional feature extraction network and Transformer module. Firstly, an improved DETR network main feature extraction module adopts the ResNet18 network and replaces some residual layers with deformable convolutions (DCNv2), enabling the model to better adapt to pollution-free fruit changes at different scales and angles, while eliminating the impact of microbial contamination on fruit testing; Subsequently, the extended spatial pyramid pooling model (DSPP) and multiscale residual aggregation module (FRAM) are integrated, which help reduce feature noise and minimize the loss of underlying features during the feature extraction process. The fusion of the two modules enhances the model’s ability to detect objects of different scales, thereby improving the accuracy of near-color fruit detection. At the same time, in order to solve the problems of slow convergence speed and large calculation amount of the basic network model, the convergence speed of the overall network model is improved by replacing the attention mechanism of Transformer. Experimental results show that compared with the original DETR model, the proposed algorithm has improved in AP, AP50, and AP75 indicators, especially in the AP50 indicator, which has the most obvious improvement reaching a detection accuracy of 97.12%. In the meantime, the trained network model is deployed on the picking robot. Compared with the original DETR network model, its average detection accuracy is as high as 96.58%, and the detection speed is increased by about 51%. Mixed sample detection tests were carried out before and after the model deployment, and the detection rate of the proposed method for nonpolluted fruits reached more than 0.95. enabling the picking robot to efficiently complete the task of picking green apples. The test results show that the algorithm proposed in this article exhibits great potential in the task of detecting pollution-free near-color fruits by the picking robot. It ensures pollution-free fruit picking and the application of AI in food safety.http://www.sciencedirect.com/science/article/pii/S0362028X24001819DETRNear color systemResNet18Sample analysisTarget detectionTransformer |
| spellingShingle | Wei Ji Kelong Zhai Bo Xu Jiawen Wu Green Apple Detection Method Based on Multidimensional Feature Extraction Network Model and Transformer Module Journal of Food Protection DETR Near color system ResNet18 Sample analysis Target detection Transformer |
| title | Green Apple Detection Method Based on Multidimensional Feature Extraction Network Model and Transformer Module |
| title_full | Green Apple Detection Method Based on Multidimensional Feature Extraction Network Model and Transformer Module |
| title_fullStr | Green Apple Detection Method Based on Multidimensional Feature Extraction Network Model and Transformer Module |
| title_full_unstemmed | Green Apple Detection Method Based on Multidimensional Feature Extraction Network Model and Transformer Module |
| title_short | Green Apple Detection Method Based on Multidimensional Feature Extraction Network Model and Transformer Module |
| title_sort | green apple detection method based on multidimensional feature extraction network model and transformer module |
| topic | DETR Near color system ResNet18 Sample analysis Target detection Transformer |
| url | http://www.sciencedirect.com/science/article/pii/S0362028X24001819 |
| work_keys_str_mv | AT weiji greenappledetectionmethodbasedonmultidimensionalfeatureextractionnetworkmodelandtransformermodule AT kelongzhai greenappledetectionmethodbasedonmultidimensionalfeatureextractionnetworkmodelandtransformermodule AT boxu greenappledetectionmethodbasedonmultidimensionalfeatureextractionnetworkmodelandtransformermodule AT jiawenwu greenappledetectionmethodbasedonmultidimensionalfeatureextractionnetworkmodelandtransformermodule |