Broad Dataset and Methods for Counting and Localization of On-Ear Corn Kernels
Crop monitoring and yield prediction are central to management decisions for farmers. One key task is counting the number of kernels on an ear of corn to estimate yield in a field. As ears of corn can easily have 400–900 kernels, manual counting is unrealistic; traditionally, growers have approximat...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-05-01
|
| Series: | Frontiers in Robotics and AI |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/frobt.2021.627009/full |
| _version_ | 1819135974499680256 |
|---|---|
| author | Jennifer Hobbs Vachik Khachatryan Barathwaj S. Anandan Barathwaj S. Anandan Harutyun Hovhannisyan David Wilson |
| author_facet | Jennifer Hobbs Vachik Khachatryan Barathwaj S. Anandan Barathwaj S. Anandan Harutyun Hovhannisyan David Wilson |
| author_sort | Jennifer Hobbs |
| collection | DOAJ |
| description | Crop monitoring and yield prediction are central to management decisions for farmers. One key task is counting the number of kernels on an ear of corn to estimate yield in a field. As ears of corn can easily have 400–900 kernels, manual counting is unrealistic; traditionally, growers have approximated the number of kernels on an ear of corn through a mixture of counting and estimation. With the success of deep learning, these human estimates can now be replaced with more accurate machine learning models, many of which are efficient enough to run on a mobile device. Although a conceptually simple task, the counting and localization of hundreds of instances in an image is challenging for many image detection algorithms which struggle when objects are small in size and large in number. We compare different detection-based frameworks, Faster R-CNN, YOLO, and density-estimation approaches for on-ear corn kernel counting and localization. In addition to the YOLOv5 model which is accurate and edge-deployable, our density-estimation approach produces high-quality results, is lightweight enough for edge deployment, and maintains its computational efficiency independent of the number of kernels in the image. Additionally, we seek to standardize and broaden this line of work through the release of a challenging dataset with high-quality, multi-class segmentation masks. This dataset firstly enables quantitative comparison of approaches within the kernel counting application space and secondly promotes further research in transfer learning and domain adaptation, large count segmentation methods, and edge deployment methods. |
| first_indexed | 2024-12-22T10:27:36Z |
| format | Article |
| id | doaj.art-a3e71e5d96794ee5a1da3491997e4842 |
| institution | Directory Open Access Journal |
| issn | 2296-9144 |
| language | English |
| last_indexed | 2024-12-22T10:27:36Z |
| publishDate | 2021-05-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Robotics and AI |
| spelling | doaj.art-a3e71e5d96794ee5a1da3491997e48422022-12-21T18:29:25ZengFrontiers Media S.A.Frontiers in Robotics and AI2296-91442021-05-01810.3389/frobt.2021.627009627009Broad Dataset and Methods for Counting and Localization of On-Ear Corn KernelsJennifer Hobbs0Vachik Khachatryan1Barathwaj S. Anandan2Barathwaj S. Anandan3Harutyun Hovhannisyan4David Wilson5Intelinair, Inc., Champaign, IL, United StatesIntelinair, Inc., Yerevan, ArmeniaIntelinair, Inc., Champaign, IL, United StatesThe Robotics Institute, Carnegie Mellon University, Pittsburgh, PA, United StatesIntelinair, Inc., Yerevan, ArmeniaIntelinair, Inc., Yerevan, ArmeniaCrop monitoring and yield prediction are central to management decisions for farmers. One key task is counting the number of kernels on an ear of corn to estimate yield in a field. As ears of corn can easily have 400–900 kernels, manual counting is unrealistic; traditionally, growers have approximated the number of kernels on an ear of corn through a mixture of counting and estimation. With the success of deep learning, these human estimates can now be replaced with more accurate machine learning models, many of which are efficient enough to run on a mobile device. Although a conceptually simple task, the counting and localization of hundreds of instances in an image is challenging for many image detection algorithms which struggle when objects are small in size and large in number. We compare different detection-based frameworks, Faster R-CNN, YOLO, and density-estimation approaches for on-ear corn kernel counting and localization. In addition to the YOLOv5 model which is accurate and edge-deployable, our density-estimation approach produces high-quality results, is lightweight enough for edge deployment, and maintains its computational efficiency independent of the number of kernels in the image. Additionally, we seek to standardize and broaden this line of work through the release of a challenging dataset with high-quality, multi-class segmentation masks. This dataset firstly enables quantitative comparison of approaches within the kernel counting application space and secondly promotes further research in transfer learning and domain adaptation, large count segmentation methods, and edge deployment methods.https://www.frontiersin.org/articles/10.3389/frobt.2021.627009/fullcountingdensity estimationprecision agriculturedatasetYOLOmachine vision application |
| spellingShingle | Jennifer Hobbs Vachik Khachatryan Barathwaj S. Anandan Barathwaj S. Anandan Harutyun Hovhannisyan David Wilson Broad Dataset and Methods for Counting and Localization of On-Ear Corn Kernels Frontiers in Robotics and AI counting density estimation precision agriculture dataset YOLO machine vision application |
| title | Broad Dataset and Methods for Counting and Localization of On-Ear Corn Kernels |
| title_full | Broad Dataset and Methods for Counting and Localization of On-Ear Corn Kernels |
| title_fullStr | Broad Dataset and Methods for Counting and Localization of On-Ear Corn Kernels |
| title_full_unstemmed | Broad Dataset and Methods for Counting and Localization of On-Ear Corn Kernels |
| title_short | Broad Dataset and Methods for Counting and Localization of On-Ear Corn Kernels |
| title_sort | broad dataset and methods for counting and localization of on ear corn kernels |
| topic | counting density estimation precision agriculture dataset YOLO machine vision application |
| url | https://www.frontiersin.org/articles/10.3389/frobt.2021.627009/full |
| work_keys_str_mv | AT jenniferhobbs broaddatasetandmethodsforcountingandlocalizationofonearcornkernels AT vachikkhachatryan broaddatasetandmethodsforcountingandlocalizationofonearcornkernels AT barathwajsanandan broaddatasetandmethodsforcountingandlocalizationofonearcornkernels AT barathwajsanandan broaddatasetandmethodsforcountingandlocalizationofonearcornkernels AT harutyunhovhannisyan broaddatasetandmethodsforcountingandlocalizationofonearcornkernels AT davidwilson broaddatasetandmethodsforcountingandlocalizationofonearcornkernels |