PocketMaize: An Android-Smartphone Application for Maize Plant Phenotyping
A low-cost portable wild phenotyping system is useful for breeders to obtain detailed phenotypic characterization to identify promising wild species. However, compared with the larger, faster, and more advanced in-laboratory phenotyping systems developed in recent years, the progress for smaller phe...
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Frontiers Media S.A.
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/2413de0b4d20492ba22e912813f66449 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| id |
oai:doaj.org-article:2413de0b4d20492ba22e912813f66449 |
|---|---|
| record_format |
dspace |
| spelling |
oai:doaj.org-article:2413de0b4d20492ba22e912813f664492021-12-01T02:29:35ZPocketMaize: An Android-Smartphone Application for Maize Plant Phenotyping1664-462X10.3389/fpls.2021.770217https://doaj.org/article/2413de0b4d20492ba22e912813f664492021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fpls.2021.770217/fullhttps://doaj.org/toc/1664-462XA low-cost portable wild phenotyping system is useful for breeders to obtain detailed phenotypic characterization to identify promising wild species. However, compared with the larger, faster, and more advanced in-laboratory phenotyping systems developed in recent years, the progress for smaller phenotyping systems, which provide fast deployment and potential for wide usage in rural and wild areas, is quite limited. In this study, we developed a portable whole-plant on-device phenotyping smartphone application running on Android that can measure up to 45 traits, including 15 plant traits, 25 leaf traits and 5 stem traits, based on images. To avoid the influence of outdoor environments, we trained a DeepLabV3+ model for segmentation. In addition, an angle calibration algorithm was also designed to reduce the error introduced by the different imaging angles. The average execution time for the analysis of a 20-million-pixel image is within 2,500 ms. The application is a portable on-device fast phenotyping platform providing methods for real-time trait measurement, which will facilitate maize phenotyping in field and benefit crop breeding in future.Lingbo LiuLejun YuLejun YuDan WuJunli YeHui FengHui FengQian LiuQian LiuWanneng YangWanneng YangFrontiers Media S.A.articlesmartphoneapplicationplant phenotypingdeep learningmaize plantsPlant cultureSB1-1110ENFrontiers in Plant Science, Vol 12 (2021) |
| institution |
DOAJ |
| collection |
DOAJ |
| language |
EN |
| topic |
smartphone application plant phenotyping deep learning maize plants Plant culture SB1-1110 |
| spellingShingle |
smartphone application plant phenotyping deep learning maize plants Plant culture SB1-1110 Lingbo Liu Lejun Yu Lejun Yu Dan Wu Junli Ye Hui Feng Hui Feng Qian Liu Qian Liu Wanneng Yang Wanneng Yang PocketMaize: An Android-Smartphone Application for Maize Plant Phenotyping |
| description |
A low-cost portable wild phenotyping system is useful for breeders to obtain detailed phenotypic characterization to identify promising wild species. However, compared with the larger, faster, and more advanced in-laboratory phenotyping systems developed in recent years, the progress for smaller phenotyping systems, which provide fast deployment and potential for wide usage in rural and wild areas, is quite limited. In this study, we developed a portable whole-plant on-device phenotyping smartphone application running on Android that can measure up to 45 traits, including 15 plant traits, 25 leaf traits and 5 stem traits, based on images. To avoid the influence of outdoor environments, we trained a DeepLabV3+ model for segmentation. In addition, an angle calibration algorithm was also designed to reduce the error introduced by the different imaging angles. The average execution time for the analysis of a 20-million-pixel image is within 2,500 ms. The application is a portable on-device fast phenotyping platform providing methods for real-time trait measurement, which will facilitate maize phenotyping in field and benefit crop breeding in future. |
| format |
article |
| author |
Lingbo Liu Lejun Yu Lejun Yu Dan Wu Junli Ye Hui Feng Hui Feng Qian Liu Qian Liu Wanneng Yang Wanneng Yang |
| author_facet |
Lingbo Liu Lejun Yu Lejun Yu Dan Wu Junli Ye Hui Feng Hui Feng Qian Liu Qian Liu Wanneng Yang Wanneng Yang |
| author_sort |
Lingbo Liu |
| title |
PocketMaize: An Android-Smartphone Application for Maize Plant Phenotyping |
| title_short |
PocketMaize: An Android-Smartphone Application for Maize Plant Phenotyping |
| title_full |
PocketMaize: An Android-Smartphone Application for Maize Plant Phenotyping |
| title_fullStr |
PocketMaize: An Android-Smartphone Application for Maize Plant Phenotyping |
| title_full_unstemmed |
PocketMaize: An Android-Smartphone Application for Maize Plant Phenotyping |
| title_sort |
pocketmaize: an android-smartphone application for maize plant phenotyping |
| publisher |
Frontiers Media S.A. |
| publishDate |
2021 |
| url |
https://doaj.org/article/2413de0b4d20492ba22e912813f66449 |
| work_keys_str_mv |
AT lingboliu pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT lejunyu pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT lejunyu pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT danwu pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT junliye pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT huifeng pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT huifeng pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT qianliu pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT qianliu pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT wannengyang pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping AT wannengyang pocketmaizeanandroidsmartphoneapplicationformaizeplantphenotyping |
| _version_ |
1718405891713138688 |