Real-Time Detection of Vine Trunk for Robot Localization Using Deep Learning Models Developed for Edge TPU Devices
The concept of the Internet of Things (IoT) in agriculture is associated with the use of high-tech devices such as robots and sensors that are interconnected to assess or monitor conditions on a particular plot of land and then deploy the various factors of production such as seeds, fertilizer, wate...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-06-01
|
| Series: | Future Internet |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-5903/14/7/199 |
| _version_ | 1827618932224163840 |
|---|---|
| author | Khadijeh Alibabaei Eduardo Assunção Pedro D. Gaspar Vasco N. G. J. Soares João M. L. P. Caldeira |
| author_facet | Khadijeh Alibabaei Eduardo Assunção Pedro D. Gaspar Vasco N. G. J. Soares João M. L. P. Caldeira |
| author_sort | Khadijeh Alibabaei |
| collection | DOAJ |
| description | The concept of the Internet of Things (IoT) in agriculture is associated with the use of high-tech devices such as robots and sensors that are interconnected to assess or monitor conditions on a particular plot of land and then deploy the various factors of production such as seeds, fertilizer, water, etc., accordingly. Vine trunk detection can help create an accurate map of the vineyard that the agricultural robot can rely on to safely navigate and perform a variety of agricultural tasks such as harvesting, pruning, etc. In this work, the state-of-the-art single-shot multibox detector (SSD) with MobileDet Edge TPU and MobileNet Edge TPU models as the backbone was used to detect the tree trunks in the vineyard. Compared to the SSD with MobileNet-V1, MobileNet-V2, and MobileDet as backbone, the SSD with MobileNet Edge TPU was more accurate in inference on the Raspberrypi, with almost the same inference time on the TPU. The SSD with MobileDet Edge TPU achieved the second-best accurate model. Additionally, this work examines the effects of some features, including the size of the input model, the quantity of training data, and the diversity of the training dataset. Increasing the size of the input model and the training dataset increased the performance of the model. |
| first_indexed | 2024-03-09T10:19:10Z |
| format | Article |
| id | doaj.art-4ad16a002e7741cf831bb8c0bfd10e7c |
| institution | Directory Open Access Journal |
| issn | 1999-5903 |
| language | English |
| last_indexed | 2024-03-09T10:19:10Z |
| publishDate | 2022-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Future Internet |
| spelling | doaj.art-4ad16a002e7741cf831bb8c0bfd10e7c2023-12-01T22:10:02ZengMDPI AGFuture Internet1999-59032022-06-0114719910.3390/fi14070199Real-Time Detection of Vine Trunk for Robot Localization Using Deep Learning Models Developed for Edge TPU DevicesKhadijeh Alibabaei0Eduardo Assunção1Pedro D. Gaspar2Vasco N. G. J. Soares3João M. L. P. Caldeira4Department of Electromechanical Engineering, University of Beira Interior, 6201-001 Covilhã, PortugalDepartment of Electromechanical Engineering, University of Beira Interior, 6201-001 Covilhã, PortugalDepartment of Electromechanical Engineering, University of Beira Interior, 6201-001 Covilhã, PortugalPolytechnic Institute of Castelo Branco, 6000-084 Castelo Branco, PortugalPolytechnic Institute of Castelo Branco, 6000-084 Castelo Branco, PortugalThe concept of the Internet of Things (IoT) in agriculture is associated with the use of high-tech devices such as robots and sensors that are interconnected to assess or monitor conditions on a particular plot of land and then deploy the various factors of production such as seeds, fertilizer, water, etc., accordingly. Vine trunk detection can help create an accurate map of the vineyard that the agricultural robot can rely on to safely navigate and perform a variety of agricultural tasks such as harvesting, pruning, etc. In this work, the state-of-the-art single-shot multibox detector (SSD) with MobileDet Edge TPU and MobileNet Edge TPU models as the backbone was used to detect the tree trunks in the vineyard. Compared to the SSD with MobileNet-V1, MobileNet-V2, and MobileDet as backbone, the SSD with MobileNet Edge TPU was more accurate in inference on the Raspberrypi, with almost the same inference time on the TPU. The SSD with MobileDet Edge TPU achieved the second-best accurate model. Additionally, this work examines the effects of some features, including the size of the input model, the quantity of training data, and the diversity of the training dataset. Increasing the size of the input model and the training dataset increased the performance of the model.https://www.mdpi.com/1999-5903/14/7/199agriculturedeep learningIoTrobottrunk detection |
| spellingShingle | Khadijeh Alibabaei Eduardo Assunção Pedro D. Gaspar Vasco N. G. J. Soares João M. L. P. Caldeira Real-Time Detection of Vine Trunk for Robot Localization Using Deep Learning Models Developed for Edge TPU Devices Future Internet agriculture deep learning IoT robot trunk detection |
| title | Real-Time Detection of Vine Trunk for Robot Localization Using Deep Learning Models Developed for Edge TPU Devices |
| title_full | Real-Time Detection of Vine Trunk for Robot Localization Using Deep Learning Models Developed for Edge TPU Devices |
| title_fullStr | Real-Time Detection of Vine Trunk for Robot Localization Using Deep Learning Models Developed for Edge TPU Devices |
| title_full_unstemmed | Real-Time Detection of Vine Trunk for Robot Localization Using Deep Learning Models Developed for Edge TPU Devices |
| title_short | Real-Time Detection of Vine Trunk for Robot Localization Using Deep Learning Models Developed for Edge TPU Devices |
| title_sort | real time detection of vine trunk for robot localization using deep learning models developed for edge tpu devices |
| topic | agriculture deep learning IoT robot trunk detection |
| url | https://www.mdpi.com/1999-5903/14/7/199 |
| work_keys_str_mv | AT khadijehalibabaei realtimedetectionofvinetrunkforrobotlocalizationusingdeeplearningmodelsdevelopedforedgetpudevices AT eduardoassuncao realtimedetectionofvinetrunkforrobotlocalizationusingdeeplearningmodelsdevelopedforedgetpudevices AT pedrodgaspar realtimedetectionofvinetrunkforrobotlocalizationusingdeeplearningmodelsdevelopedforedgetpudevices AT vascongjsoares realtimedetectionofvinetrunkforrobotlocalizationusingdeeplearningmodelsdevelopedforedgetpudevices AT joaomlpcaldeira realtimedetectionofvinetrunkforrobotlocalizationusingdeeplearningmodelsdevelopedforedgetpudevices |