Super resolution for root imaging
Premise High‐resolution cameras are very helpful for plant phenotyping as their images enable tasks such as target vs. background discrimination and the measurement and analysis of fine above‐ground plant attributes. However, the acquisition of high‐resolution images of plant roots is more challengi...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2020-07-01
|
| Series: | Applications in Plant Sciences |
| Subjects: | |
| Online Access: | https://doi.org/10.1002/aps3.11374 |
| _version_ | 1818329090860515328 |
|---|---|
| author | Jose F. Ruiz‐Munoz Jyothier K. Nimmagadda Tyler G. Dowd James E. Baciak Alina Zare |
| author_facet | Jose F. Ruiz‐Munoz Jyothier K. Nimmagadda Tyler G. Dowd James E. Baciak Alina Zare |
| author_sort | Jose F. Ruiz‐Munoz |
| collection | DOAJ |
| description | Premise High‐resolution cameras are very helpful for plant phenotyping as their images enable tasks such as target vs. background discrimination and the measurement and analysis of fine above‐ground plant attributes. However, the acquisition of high‐resolution images of plant roots is more challenging than above‐ground data collection. An effective super‐resolution (SR) algorithm is therefore needed for overcoming the resolution limitations of sensors, reducing storage space requirements, and boosting the performance of subsequent analyses. Methods We propose an SR framework for enhancing images of plant roots using convolutional neural networks. We compare three alternatives for training the SR model: (i) training with non‐plant‐root images, (ii) training with plant‐root images, and (iii) pretraining the model with non‐plant‐root images and fine‐tuning with plant‐root images. The architectures of the SR models were based on two state‐of‐the‐art deep learning approaches: a fast SR convolutional neural network and an SR generative adversarial network. Results In our experiments, we observed that the SR models improved the quality of low‐resolution images of plant roots in an unseen data set in terms of the signal‐to‐noise ratio. We used a collection of publicly available data sets to demonstrate that the SR models outperform the basic bicubic interpolation, even when trained with non‐root data sets. Discussion The incorporation of a deep learning–based SR model in the imaging process enhances the quality of low‐resolution images of plant roots. We demonstrate that SR preprocessing boosts the performance of a machine learning system trained to separate plant roots from their background. Our segmentation experiments also show that high performance on this task can be achieved independently of the signal‐to‐noise ratio. We therefore conclude that the quality of the image enhancement depends on the desired application. |
| first_indexed | 2024-12-13T12:42:32Z |
| format | Article |
| id | doaj.art-869d185355ed43f9bf493ddccc889454 |
| institution | Directory Open Access Journal |
| issn | 2168-0450 |
| language | English |
| last_indexed | 2024-12-13T12:42:32Z |
| publishDate | 2020-07-01 |
| publisher | Wiley |
| record_format | Article |
| series | Applications in Plant Sciences |
| spelling | doaj.art-869d185355ed43f9bf493ddccc8894542022-12-21T23:45:36ZengWileyApplications in Plant Sciences2168-04502020-07-0187n/an/a10.1002/aps3.11374Super resolution for root imagingJose F. Ruiz‐Munoz0Jyothier K. Nimmagadda1Tyler G. Dowd2James E. Baciak3Alina Zare4Department of Electrical and Computer Engineering University of Florida Gainesville Florida USADepartment of Material Sciences and Engineering University of Florida Gainesville Florida USADonald Danforth Plant Science Center St. Louis Missouri USADepartment of Material Sciences and Engineering University of Florida Gainesville Florida USADepartment of Electrical and Computer Engineering University of Florida Gainesville Florida USAPremise High‐resolution cameras are very helpful for plant phenotyping as their images enable tasks such as target vs. background discrimination and the measurement and analysis of fine above‐ground plant attributes. However, the acquisition of high‐resolution images of plant roots is more challenging than above‐ground data collection. An effective super‐resolution (SR) algorithm is therefore needed for overcoming the resolution limitations of sensors, reducing storage space requirements, and boosting the performance of subsequent analyses. Methods We propose an SR framework for enhancing images of plant roots using convolutional neural networks. We compare three alternatives for training the SR model: (i) training with non‐plant‐root images, (ii) training with plant‐root images, and (iii) pretraining the model with non‐plant‐root images and fine‐tuning with plant‐root images. The architectures of the SR models were based on two state‐of‐the‐art deep learning approaches: a fast SR convolutional neural network and an SR generative adversarial network. Results In our experiments, we observed that the SR models improved the quality of low‐resolution images of plant roots in an unseen data set in terms of the signal‐to‐noise ratio. We used a collection of publicly available data sets to demonstrate that the SR models outperform the basic bicubic interpolation, even when trained with non‐root data sets. Discussion The incorporation of a deep learning–based SR model in the imaging process enhances the quality of low‐resolution images of plant roots. We demonstrate that SR preprocessing boosts the performance of a machine learning system trained to separate plant roots from their background. Our segmentation experiments also show that high performance on this task can be achieved independently of the signal‐to‐noise ratio. We therefore conclude that the quality of the image enhancement depends on the desired application.https://doi.org/10.1002/aps3.11374convolutional neural networksgenerative adversarial networksplant phenotypingroot phenotypingsuper resolution |
| spellingShingle | Jose F. Ruiz‐Munoz Jyothier K. Nimmagadda Tyler G. Dowd James E. Baciak Alina Zare Super resolution for root imaging Applications in Plant Sciences convolutional neural networks generative adversarial networks plant phenotyping root phenotyping super resolution |
| title | Super resolution for root imaging |
| title_full | Super resolution for root imaging |
| title_fullStr | Super resolution for root imaging |
| title_full_unstemmed | Super resolution for root imaging |
| title_short | Super resolution for root imaging |
| title_sort | super resolution for root imaging |
| topic | convolutional neural networks generative adversarial networks plant phenotyping root phenotyping super resolution |
| url | https://doi.org/10.1002/aps3.11374 |
| work_keys_str_mv | AT josefruizmunoz superresolutionforrootimaging AT jyothierknimmagadda superresolutionforrootimaging AT tylergdowd superresolutionforrootimaging AT jamesebaciak superresolutionforrootimaging AT alinazare superresolutionforrootimaging |