Chrysanthemum classification method integrating deep visual features from both the front and back sides
IntroducionChrysanthemum morifolium Ramat (hereinafter referred to as Chrysanthemum) is one of the most beloved and economically valuable Chinese herbal crops, which contains abundant medicinal ingredients and wide application prospects. Therefore, identifying the classification and origin of Chrysa...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-01-01
|
| Series: | Frontiers in Plant Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpls.2024.1463113/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832592429456293888 |
|---|---|
| author | Yifan Chen Xichen Yang Hui Yan Hui Yan Jia Liu Jian Jiang Zhongyuan Mao Tianshu Wang Tianshu Wang |
| author_facet | Yifan Chen Xichen Yang Hui Yan Hui Yan Jia Liu Jian Jiang Zhongyuan Mao Tianshu Wang Tianshu Wang |
| author_sort | Yifan Chen |
| collection | DOAJ |
| description | IntroducionChrysanthemum morifolium Ramat (hereinafter referred to as Chrysanthemum) is one of the most beloved and economically valuable Chinese herbal crops, which contains abundant medicinal ingredients and wide application prospects. Therefore, identifying the classification and origin of Chrysanthemum is important for producers, consumers, and market regulators. The existing Chrysanthemum classification methods mostly rely on visual subjective identification, are time-consuming, and always need high equipment costs.MethodsA novel method is proposed to accurately identify the Chrysanthemum classification in a swift, non-invasive, and non-contact way. The proposed method is based on the fusion of deep visual features of both the front and back sides. Firstly, the different Chrysanthemums images are collected and labeled with origins and classifications. Secondly, the background area with less available information is removed by image preprocessing. Thirdly, a two-stream feature extraction network is designed with two inputs which are the preprocessed front and back Chrysanthemum images. Meanwhile, the incorporation of single-stream residual connections and cross-stream residual connections is employed to extend the receptive field of the network and fully fusion the features from both the front and back sides.ResultsExperimental results demonstrate that the proposed method achieves an accuracy of 93.8%, outperforming existing methods and exhibiting superior stability.DiscussionThe proposed method provides an effective and dependable solution for identifying Chrysanthemum classification and origin while offering practical benefits for quality assurance in production, consumer markets, and regulatory processes. Code and data are available at https://github.com/dart-into/CCMIFB. |
| format | Article |
| id | doaj-art-b311b1f9a81e42dd97393b47883814e2 |
| institution | Kabale University |
| issn | 1664-462X |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Plant Science |
| spelling | doaj-art-b311b1f9a81e42dd97393b47883814e22025-01-21T08:36:48ZengFrontiers Media S.A.Frontiers in Plant Science1664-462X2025-01-011510.3389/fpls.2024.14631131463113Chrysanthemum classification method integrating deep visual features from both the front and back sidesYifan Chen0Xichen Yang1Hui Yan2Hui Yan3Jia Liu4Jian Jiang5Zhongyuan Mao6Tianshu Wang7Tianshu Wang8School of Computer and Electronic Information/School of Artificial Intelligence, Nanjing Normal University, Nanjing, Jiangsu, ChinaSchool of Computer and Electronic Information/School of Artificial Intelligence, Nanjing Normal University, Nanjing, Jiangsu, ChinaNanjing University of Chinese Medicine, National and Local Collaborative Engineering Center of Chinese Medicinal Resources Industrialization and Formulae Innovative Medicine, Nanjing, ChinaJiangsu Collaborative Innovation Center of Chinese Medicinal Resources Industrialization, Nanjing University of Chinese Medicine, Nanjing, Jiangsu, ChinaCollege of Artificial Intelligence and Information Technology, Nanjing University of Chinese Medicine, Nanjing, Jiangsu, ChinaSchool of Computer and Electronic Information/School of Artificial Intelligence, Nanjing Normal University, Nanjing, Jiangsu, ChinaSchool of Computer and Electronic Information/School of Artificial Intelligence, Nanjing Normal University, Nanjing, Jiangsu, ChinaCollege of Artificial Intelligence and Information Technology, Nanjing University of Chinese Medicine, Nanjing, Jiangsu, ChinaJiangsu Province Engineering Research Center of Traditional Chinese Medicine (TCM) Intelligence Health Service, Nanjing University of Chinese Medicine, Nanjing, Jiangsu, ChinaIntroducionChrysanthemum morifolium Ramat (hereinafter referred to as Chrysanthemum) is one of the most beloved and economically valuable Chinese herbal crops, which contains abundant medicinal ingredients and wide application prospects. Therefore, identifying the classification and origin of Chrysanthemum is important for producers, consumers, and market regulators. The existing Chrysanthemum classification methods mostly rely on visual subjective identification, are time-consuming, and always need high equipment costs.MethodsA novel method is proposed to accurately identify the Chrysanthemum classification in a swift, non-invasive, and non-contact way. The proposed method is based on the fusion of deep visual features of both the front and back sides. Firstly, the different Chrysanthemums images are collected and labeled with origins and classifications. Secondly, the background area with less available information is removed by image preprocessing. Thirdly, a two-stream feature extraction network is designed with two inputs which are the preprocessed front and back Chrysanthemum images. Meanwhile, the incorporation of single-stream residual connections and cross-stream residual connections is employed to extend the receptive field of the network and fully fusion the features from both the front and back sides.ResultsExperimental results demonstrate that the proposed method achieves an accuracy of 93.8%, outperforming existing methods and exhibiting superior stability.DiscussionThe proposed method provides an effective and dependable solution for identifying Chrysanthemum classification and origin while offering practical benefits for quality assurance in production, consumer markets, and regulatory processes. Code and data are available at https://github.com/dart-into/CCMIFB.https://www.frontiersin.org/articles/10.3389/fpls.2024.1463113/fullChrysanthemum classificationtwo-stream networkvisual informationfeature fusiondeep learning |
| spellingShingle | Yifan Chen Xichen Yang Hui Yan Hui Yan Jia Liu Jian Jiang Zhongyuan Mao Tianshu Wang Tianshu Wang Chrysanthemum classification method integrating deep visual features from both the front and back sides Frontiers in Plant Science Chrysanthemum classification two-stream network visual information feature fusion deep learning |
| title | Chrysanthemum classification method integrating deep visual features from both the front and back sides |
| title_full | Chrysanthemum classification method integrating deep visual features from both the front and back sides |
| title_fullStr | Chrysanthemum classification method integrating deep visual features from both the front and back sides |
| title_full_unstemmed | Chrysanthemum classification method integrating deep visual features from both the front and back sides |
| title_short | Chrysanthemum classification method integrating deep visual features from both the front and back sides |
| title_sort | chrysanthemum classification method integrating deep visual features from both the front and back sides |
| topic | Chrysanthemum classification two-stream network visual information feature fusion deep learning |
| url | https://www.frontiersin.org/articles/10.3389/fpls.2024.1463113/full |
| work_keys_str_mv | AT yifanchen chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides AT xichenyang chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides AT huiyan chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides AT huiyan chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides AT jialiu chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides AT jianjiang chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides AT zhongyuanmao chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides AT tianshuwang chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides AT tianshuwang chrysanthemumclassificationmethodintegratingdeepvisualfeaturesfromboththefrontandbacksides |