A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification
Abstract Traditional warehousing typically needs machine learning or manual tagging to classify objects. However, this method is less robust and consumes a lot of labour and material resources. Based on DenseNet, this work proposes a feature weighting convolutional network recognition model and desi...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2024-02-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12971 |
| _version_ | 1797311362599223296 |
|---|---|
| author | Kang An Xiaoqing Sun Yaqing Song Yebin Lu Qianqian Shangguan |
| author_facet | Kang An Xiaoqing Sun Yaqing Song Yebin Lu Qianqian Shangguan |
| author_sort | Kang An |
| collection | DOAJ |
| description | Abstract Traditional warehousing typically needs machine learning or manual tagging to classify objects. However, this method is less robust and consumes a lot of labour and material resources. Based on DenseNet, this work proposes a feature weighting convolutional network recognition model and designs a set of software and hardware for data acquisition, which is applied to the efficient classification of industrial parts in warehouse management. Firstly, this work modifies DenseNet by embedding SE‐Block, and replaces the cross‐entropy loss function with the focus loss function to optimize the model structure. Secondly, a multi‐view hardware and software acquisition system is designed to complete the functions of part image acquisition, image preprocessing, model training and part recognition. Finally, an industrial parts sorting experiment was designed. Compared with the original DenseNet model, the proposed weighted convolutional network identification model showed that the accuracy of the modified model was increased by 3.09% and the convergence rate was significantly improved. The modified model proposed in this work aims to improve the recognition accuracy of industrial parts in modern warehouse management, so as to modify the classification efficiency of warehouse parts in production. |
| first_indexed | 2024-03-08T01:57:33Z |
| format | Article |
| id | doaj.art-185eab480a7d4e1b99fb921633ca77c7 |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-03-08T01:57:33Z |
| publishDate | 2024-02-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-185eab480a7d4e1b99fb921633ca77c72024-02-14T07:53:24ZengWileyIET Image Processing1751-96591751-96672024-02-0118358960110.1049/ipr2.12971A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classificationKang An0Xiaoqing Sun1Yaqing Song2Yebin Lu3Qianqian Shangguan4College of Information, Mechanical and Electrical Engineering Shanghai Normal University Shanghai ChinaCollege of Information, Mechanical and Electrical Engineering Shanghai Normal University Shanghai ChinaCollege of Information, Mechanical and Electrical Engineering Shanghai Normal University Shanghai ChinaCollege of Information, Mechanical and Electrical Engineering Shanghai Normal University Shanghai ChinaCollege of Information, Mechanical and Electrical Engineering Shanghai Normal University Shanghai ChinaAbstract Traditional warehousing typically needs machine learning or manual tagging to classify objects. However, this method is less robust and consumes a lot of labour and material resources. Based on DenseNet, this work proposes a feature weighting convolutional network recognition model and designs a set of software and hardware for data acquisition, which is applied to the efficient classification of industrial parts in warehouse management. Firstly, this work modifies DenseNet by embedding SE‐Block, and replaces the cross‐entropy loss function with the focus loss function to optimize the model structure. Secondly, a multi‐view hardware and software acquisition system is designed to complete the functions of part image acquisition, image preprocessing, model training and part recognition. Finally, an industrial parts sorting experiment was designed. Compared with the original DenseNet model, the proposed weighted convolutional network identification model showed that the accuracy of the modified model was increased by 3.09% and the convergence rate was significantly improved. The modified model proposed in this work aims to improve the recognition accuracy of industrial parts in modern warehouse management, so as to modify the classification efficiency of warehouse parts in production.https://doi.org/10.1049/ipr2.12971character recognitionconvolutional neural netsedge detectionlearning (artificial intelligence) |
| spellingShingle | Kang An Xiaoqing Sun Yaqing Song Yebin Lu Qianqian Shangguan A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification IET Image Processing character recognition convolutional neural nets edge detection learning (artificial intelligence) |
| title | A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification |
| title_full | A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification |
| title_fullStr | A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification |
| title_full_unstemmed | A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification |
| title_short | A DenseNet‐based feature weighting convolutional network recognition model and its application in industrial part classification |
| title_sort | densenet based feature weighting convolutional network recognition model and its application in industrial part classification |
| topic | character recognition convolutional neural nets edge detection learning (artificial intelligence) |
| url | https://doi.org/10.1049/ipr2.12971 |
| work_keys_str_mv | AT kangan adensenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT xiaoqingsun adensenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT yaqingsong adensenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT yebinlu adensenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT qianqianshangguan adensenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT kangan densenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT xiaoqingsun densenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT yaqingsong densenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT yebinlu densenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification AT qianqianshangguan densenetbasedfeatureweightingconvolutionalnetworkrecognitionmodelanditsapplicationinindustrialpartclassification |