Counting Dense Leaves under Natural Environments via an Improved Deep-Learning-Based Object Detection Algorithm
The leaf is the organ that is crucial for photosynthesis and the production of nutrients in plants; as such, the number of leaves is one of the key indicators with which to describe the development and growth of a canopy. The irregular shape and distribution of the blades, as well as the effect of n...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-10-01
|
| Series: | Agriculture |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2077-0472/11/10/1003 |
| _version_ | 1827600666573406208 |
|---|---|
| author | Shenglian Lu Zhen Song Wenkang Chen Tingting Qian Yingyu Zhang Ming Chen Guo Li |
| author_facet | Shenglian Lu Zhen Song Wenkang Chen Tingting Qian Yingyu Zhang Ming Chen Guo Li |
| author_sort | Shenglian Lu |
| collection | DOAJ |
| description | The leaf is the organ that is crucial for photosynthesis and the production of nutrients in plants; as such, the number of leaves is one of the key indicators with which to describe the development and growth of a canopy. The irregular shape and distribution of the blades, as well as the effect of natural light, make the segmentation and detection process of the blades difficult. The inaccurate acquisition of plant phenotypic parameters may affect the subsequent judgment of crop growth status and crop yield. To address the challenge in counting dense and overlapped plant leaves under natural environments, we proposed an improved deep-learning-based object detection algorithm by merging a space-to-depth module, a Convolutional Block Attention Module (CBAM) and Atrous Spatial Pyramid Pooling (ASPP) into the network, and applying the <i>smooth<sub>L1</sub></i> function to improve the loss function of object prediction. We evaluated our method on images of five different plant species collected under indoor and outdoor environments. The experimental results demonstrated that our algorithm which counts dense leaves improved average detection accuracy of 85% to 96%. Our algorithm also showed better performance in both detection accuracy and time consumption compared to other state-of-the-art object detection algorithms. |
| first_indexed | 2024-03-09T04:40:24Z |
| format | Article |
| id | doaj.art-ce8e5db5327642c2b8af6f60bd2f5292 |
| institution | Directory Open Access Journal |
| issn | 2077-0472 |
| language | English |
| last_indexed | 2024-03-09T04:40:24Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Agriculture |
| spelling | doaj.art-ce8e5db5327642c2b8af6f60bd2f52922023-12-03T13:22:26ZengMDPI AGAgriculture2077-04722021-10-011110100310.3390/agriculture11101003Counting Dense Leaves under Natural Environments via an Improved Deep-Learning-Based Object Detection AlgorithmShenglian Lu0Zhen Song1Wenkang Chen2Tingting Qian3Yingyu Zhang4Ming Chen5Guo Li6Guangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaGuangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaGuangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaAgricultural Information Institutes of Science and Technology, Shanghai Academy of Agriculture Sciences, Shanghai 201403, ChinaAgricultural Information Institutes of Science and Technology, Shanghai Academy of Agriculture Sciences, Shanghai 201403, ChinaGuangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaGuangxi Key Lab of Multisource Information Mining & Security, College of Computer Science & Engineering, Guangxi Normal University, Guilin 541004, ChinaThe leaf is the organ that is crucial for photosynthesis and the production of nutrients in plants; as such, the number of leaves is one of the key indicators with which to describe the development and growth of a canopy. The irregular shape and distribution of the blades, as well as the effect of natural light, make the segmentation and detection process of the blades difficult. The inaccurate acquisition of plant phenotypic parameters may affect the subsequent judgment of crop growth status and crop yield. To address the challenge in counting dense and overlapped plant leaves under natural environments, we proposed an improved deep-learning-based object detection algorithm by merging a space-to-depth module, a Convolutional Block Attention Module (CBAM) and Atrous Spatial Pyramid Pooling (ASPP) into the network, and applying the <i>smooth<sub>L1</sub></i> function to improve the loss function of object prediction. We evaluated our method on images of five different plant species collected under indoor and outdoor environments. The experimental results demonstrated that our algorithm which counts dense leaves improved average detection accuracy of 85% to 96%. Our algorithm also showed better performance in both detection accuracy and time consumption compared to other state-of-the-art object detection algorithms.https://www.mdpi.com/2077-0472/11/10/1003deep learningplant phenotypingleaf detectionobject recognition |
| spellingShingle | Shenglian Lu Zhen Song Wenkang Chen Tingting Qian Yingyu Zhang Ming Chen Guo Li Counting Dense Leaves under Natural Environments via an Improved Deep-Learning-Based Object Detection Algorithm Agriculture deep learning plant phenotyping leaf detection object recognition |
| title | Counting Dense Leaves under Natural Environments via an Improved Deep-Learning-Based Object Detection Algorithm |
| title_full | Counting Dense Leaves under Natural Environments via an Improved Deep-Learning-Based Object Detection Algorithm |
| title_fullStr | Counting Dense Leaves under Natural Environments via an Improved Deep-Learning-Based Object Detection Algorithm |
| title_full_unstemmed | Counting Dense Leaves under Natural Environments via an Improved Deep-Learning-Based Object Detection Algorithm |
| title_short | Counting Dense Leaves under Natural Environments via an Improved Deep-Learning-Based Object Detection Algorithm |
| title_sort | counting dense leaves under natural environments via an improved deep learning based object detection algorithm |
| topic | deep learning plant phenotyping leaf detection object recognition |
| url | https://www.mdpi.com/2077-0472/11/10/1003 |
| work_keys_str_mv | AT shenglianlu countingdenseleavesundernaturalenvironmentsviaanimproveddeeplearningbasedobjectdetectionalgorithm AT zhensong countingdenseleavesundernaturalenvironmentsviaanimproveddeeplearningbasedobjectdetectionalgorithm AT wenkangchen countingdenseleavesundernaturalenvironmentsviaanimproveddeeplearningbasedobjectdetectionalgorithm AT tingtingqian countingdenseleavesundernaturalenvironmentsviaanimproveddeeplearningbasedobjectdetectionalgorithm AT yingyuzhang countingdenseleavesundernaturalenvironmentsviaanimproveddeeplearningbasedobjectdetectionalgorithm AT mingchen countingdenseleavesundernaturalenvironmentsviaanimproveddeeplearningbasedobjectdetectionalgorithm AT guoli countingdenseleavesundernaturalenvironmentsviaanimproveddeeplearningbasedobjectdetectionalgorithm |