IPMCNet: A Lightweight Algorithm for Invasive Plant Multiclassification
Invasive plant species pose significant biodiversity and ecosystem threats. Real-time identification of invasive plants is a crucial prerequisite for early and timely prevention. While deep learning has shown promising results in plant recognition, the use of deep learning models often involve a lar...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-02-01
|
| Series: | Agronomy |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4395/14/2/333 |
| _version_ | 1827344420112957440 |
|---|---|
| author | Ying Chen Xi Qiao Feng Qin Hongtao Huang Bo Liu Zaiyuan Li Conghui Liu Quan Wang Fanghao Wan Wanqiang Qian Yiqi Huang |
| author_facet | Ying Chen Xi Qiao Feng Qin Hongtao Huang Bo Liu Zaiyuan Li Conghui Liu Quan Wang Fanghao Wan Wanqiang Qian Yiqi Huang |
| author_sort | Ying Chen |
| collection | DOAJ |
| description | Invasive plant species pose significant biodiversity and ecosystem threats. Real-time identification of invasive plants is a crucial prerequisite for early and timely prevention. While deep learning has shown promising results in plant recognition, the use of deep learning models often involve a large number of parameters and high data requirements for training. Unfortunately, the available data for various invasive plant species are often limited. To address this challenge, this study proposes a lightweight deep learning model called IPMCNet for the identification of multiple invasive plant species. IPMCNet attains high recognition accuracy even with limited data and exhibits strong generalizability. Simultaneously, by employing depth-wise separable convolutional kernels, splitting channels, and eliminating fully connected layer, the model’s parameter count is lower than that of some existing lightweight models. Additionally, the study explores the impact of different loss functions, and the insertion of various attention modules on the model’s accuracy. The experimental results reveal that, compared with eight other existing neural network models, IPMCNet achieves the highest classification accuracy of 94.52%. Furthermore, the findings suggest that focal loss is the most effective loss function. The performance of the six attention modules is suboptimal, and their insertion leads to a decrease in model accuracy. |
| first_indexed | 2024-03-07T22:46:00Z |
| format | Article |
| id | doaj.art-d0b12709f36f47c7bfab4c62bcd77fa1 |
| institution | Directory Open Access Journal |
| issn | 2073-4395 |
| language | English |
| last_indexed | 2024-03-07T22:46:00Z |
| publishDate | 2024-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agronomy |
| spelling | doaj.art-d0b12709f36f47c7bfab4c62bcd77fa12024-02-23T15:04:13ZengMDPI AGAgronomy2073-43952024-02-0114233310.3390/agronomy14020333IPMCNet: A Lightweight Algorithm for Invasive Plant MulticlassificationYing Chen0Xi Qiao1Feng Qin2Hongtao Huang3Bo Liu4Zaiyuan Li5Conghui Liu6Quan Wang7Fanghao Wan8Wanqiang Qian9Yiqi Huang10College of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaCollege of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaCollege of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaCollege of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaShenzhen Branch, Guangdong Laboratory of Lingnan Modern Agriculture, Genome Analysis Laboratory of the Ministry of Agriculture and Rural Affairs, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen 518120, ChinaShenzhen Branch, Guangdong Laboratory of Lingnan Modern Agriculture, Genome Analysis Laboratory of the Ministry of Agriculture and Rural Affairs, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen 518120, ChinaShenzhen Branch, Guangdong Laboratory of Lingnan Modern Agriculture, Genome Analysis Laboratory of the Ministry of Agriculture and Rural Affairs, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen 518120, ChinaShenzhen Branch, Guangdong Laboratory of Lingnan Modern Agriculture, Genome Analysis Laboratory of the Ministry of Agriculture and Rural Affairs, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen 518120, ChinaShenzhen Branch, Guangdong Laboratory of Lingnan Modern Agriculture, Genome Analysis Laboratory of the Ministry of Agriculture and Rural Affairs, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen 518120, ChinaShenzhen Branch, Guangdong Laboratory of Lingnan Modern Agriculture, Genome Analysis Laboratory of the Ministry of Agriculture and Rural Affairs, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen 518120, ChinaCollege of Mechanical Engineering, Guangxi University, Nanning 530004, ChinaInvasive plant species pose significant biodiversity and ecosystem threats. Real-time identification of invasive plants is a crucial prerequisite for early and timely prevention. While deep learning has shown promising results in plant recognition, the use of deep learning models often involve a large number of parameters and high data requirements for training. Unfortunately, the available data for various invasive plant species are often limited. To address this challenge, this study proposes a lightweight deep learning model called IPMCNet for the identification of multiple invasive plant species. IPMCNet attains high recognition accuracy even with limited data and exhibits strong generalizability. Simultaneously, by employing depth-wise separable convolutional kernels, splitting channels, and eliminating fully connected layer, the model’s parameter count is lower than that of some existing lightweight models. Additionally, the study explores the impact of different loss functions, and the insertion of various attention modules on the model’s accuracy. The experimental results reveal that, compared with eight other existing neural network models, IPMCNet achieves the highest classification accuracy of 94.52%. Furthermore, the findings suggest that focal loss is the most effective loss function. The performance of the six attention modules is suboptimal, and their insertion leads to a decrease in model accuracy.https://www.mdpi.com/2073-4395/14/2/333deep learningplant identificationin-field detectionconvolutional neural networkinvasive plants |
| spellingShingle | Ying Chen Xi Qiao Feng Qin Hongtao Huang Bo Liu Zaiyuan Li Conghui Liu Quan Wang Fanghao Wan Wanqiang Qian Yiqi Huang IPMCNet: A Lightweight Algorithm for Invasive Plant Multiclassification Agronomy deep learning plant identification in-field detection convolutional neural network invasive plants |
| title | IPMCNet: A Lightweight Algorithm for Invasive Plant Multiclassification |
| title_full | IPMCNet: A Lightweight Algorithm for Invasive Plant Multiclassification |
| title_fullStr | IPMCNet: A Lightweight Algorithm for Invasive Plant Multiclassification |
| title_full_unstemmed | IPMCNet: A Lightweight Algorithm for Invasive Plant Multiclassification |
| title_short | IPMCNet: A Lightweight Algorithm for Invasive Plant Multiclassification |
| title_sort | ipmcnet a lightweight algorithm for invasive plant multiclassification |
| topic | deep learning plant identification in-field detection convolutional neural network invasive plants |
| url | https://www.mdpi.com/2073-4395/14/2/333 |
| work_keys_str_mv | AT yingchen ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT xiqiao ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT fengqin ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT hongtaohuang ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT boliu ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT zaiyuanli ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT conghuiliu ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT quanwang ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT fanghaowan ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT wanqiangqian ipmcnetalightweightalgorithmforinvasiveplantmulticlassification AT yiqihuang ipmcnetalightweightalgorithmforinvasiveplantmulticlassification |