Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy Scans
This study tested whether machine learning (ML) methods can effectively separate individual plants from complex 3D canopy laser scans as a prerequisite to analyzing particular plant features. For this, we scanned mung bean and chickpea crops with PlantEye (R) laser scanners. Firstly, we segmented th...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/23/8022 |
| _version_ | 1797507168201605120 |
|---|---|
| author | Serkan Kartal Sunita Choudhary Jan Masner Jana Kholová Michal Stočes Priyanka Gattu Stefan Schwartz Ewaut Kissel |
| author_facet | Serkan Kartal Sunita Choudhary Jan Masner Jana Kholová Michal Stočes Priyanka Gattu Stefan Schwartz Ewaut Kissel |
| author_sort | Serkan Kartal |
| collection | DOAJ |
| description | This study tested whether machine learning (ML) methods can effectively separate individual plants from complex 3D canopy laser scans as a prerequisite to analyzing particular plant features. For this, we scanned mung bean and chickpea crops with PlantEye (R) laser scanners. Firstly, we segmented the crop canopies from the background in 3D space using the Region Growing Segmentation algorithm. Then, Convolutional Neural Network (CNN) based ML algorithms were fine-tuned for plant counting. Application of the CNN-based (Convolutional Neural Network) processing architecture was possible only after we reduced the dimensionality of the data to 2D. This allowed for the identification of individual plants and their counting with an accuracy of 93.18% and 92.87% for mung bean and chickpea plants, respectively. These steps were connected to the phenotyping pipeline, which can now replace manual counting operations that are inefficient, costly, and error-prone. The use of CNN in this study was innovatively solved with dimensionality reduction, addition of height information as color, and consequent application of a 2D CNN-based approach. We found there to be a wide gap in the use of ML on 3D information. This gap will have to be addressed, especially for more complex plant feature extractions, which we intend to implement through further research. |
| first_indexed | 2024-03-10T04:45:45Z |
| format | Article |
| id | doaj.art-be5bb2fb943545fca868cac803dfa73d |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T04:45:45Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-be5bb2fb943545fca868cac803dfa73d2023-11-23T03:02:48ZengMDPI AGSensors1424-82202021-12-012123802210.3390/s21238022Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy ScansSerkan Kartal0Sunita Choudhary1Jan Masner2Jana Kholová3Michal Stočes4Priyanka Gattu5Stefan Schwartz6Ewaut Kissel7Department of Computer Engineering, Faculty of Engineering, Cukurova University, Adana 01330, TurkeySystem Analysis for Climate Smart Agriculture (SACSA), ISD, International Crops Research Institute for the Semi-Arid Tropics, Patancheru 5023204, Telangana, IndiaDepartment of Information Technologies, Faculty of Economics and Management, Czech University of Life Sciences Prague, Kamýcká 129, 165 00 Prague, Czech RepublicSystem Analysis for Climate Smart Agriculture (SACSA), ISD, International Crops Research Institute for the Semi-Arid Tropics, Patancheru 5023204, Telangana, IndiaDepartment of Information Technologies, Faculty of Economics and Management, Czech University of Life Sciences Prague, Kamýcká 129, 165 00 Prague, Czech RepublicDepartment of Electrical Engineering, Indian Institute of Technology Hyderabad, Sangareddy 502285, Telangana, IndiaPhenospex B. V., Jan Campertstraat 11, 6416 SG Heerlen, The NetherlandsPhenospex B. V., Jan Campertstraat 11, 6416 SG Heerlen, The NetherlandsThis study tested whether machine learning (ML) methods can effectively separate individual plants from complex 3D canopy laser scans as a prerequisite to analyzing particular plant features. For this, we scanned mung bean and chickpea crops with PlantEye (R) laser scanners. Firstly, we segmented the crop canopies from the background in 3D space using the Region Growing Segmentation algorithm. Then, Convolutional Neural Network (CNN) based ML algorithms were fine-tuned for plant counting. Application of the CNN-based (Convolutional Neural Network) processing architecture was possible only after we reduced the dimensionality of the data to 2D. This allowed for the identification of individual plants and their counting with an accuracy of 93.18% and 92.87% for mung bean and chickpea plants, respectively. These steps were connected to the phenotyping pipeline, which can now replace manual counting operations that are inefficient, costly, and error-prone. The use of CNN in this study was innovatively solved with dimensionality reduction, addition of height information as color, and consequent application of a 2D CNN-based approach. We found there to be a wide gap in the use of ML on 3D information. This gap will have to be addressed, especially for more complex plant feature extractions, which we intend to implement through further research.https://www.mdpi.com/1424-8220/21/23/80223D point cloudsplant detectionmachine learningcomputer visionphenotyping |
| spellingShingle | Serkan Kartal Sunita Choudhary Jan Masner Jana Kholová Michal Stočes Priyanka Gattu Stefan Schwartz Ewaut Kissel Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy Scans Sensors 3D point clouds plant detection machine learning computer vision phenotyping |
| title | Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy Scans |
| title_full | Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy Scans |
| title_fullStr | Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy Scans |
| title_full_unstemmed | Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy Scans |
| title_short | Machine Learning-Based Plant Detection Algorithms to Automate Counting Tasks Using 3D Canopy Scans |
| title_sort | machine learning based plant detection algorithms to automate counting tasks using 3d canopy scans |
| topic | 3D point clouds plant detection machine learning computer vision phenotyping |
| url | https://www.mdpi.com/1424-8220/21/23/8022 |
| work_keys_str_mv | AT serkankartal machinelearningbasedplantdetectionalgorithmstoautomatecountingtasksusing3dcanopyscans AT sunitachoudhary machinelearningbasedplantdetectionalgorithmstoautomatecountingtasksusing3dcanopyscans AT janmasner machinelearningbasedplantdetectionalgorithmstoautomatecountingtasksusing3dcanopyscans AT janakholova machinelearningbasedplantdetectionalgorithmstoautomatecountingtasksusing3dcanopyscans AT michalstoces machinelearningbasedplantdetectionalgorithmstoautomatecountingtasksusing3dcanopyscans AT priyankagattu machinelearningbasedplantdetectionalgorithmstoautomatecountingtasksusing3dcanopyscans AT stefanschwartz machinelearningbasedplantdetectionalgorithmstoautomatecountingtasksusing3dcanopyscans AT ewautkissel machinelearningbasedplantdetectionalgorithmstoautomatecountingtasksusing3dcanopyscans |