RootPainter:deep learning segmentation of biological images with corrective annotation
| Autor: | Jens Petersen, Eusun Han, Kristian Thorup-Kristensen, Abraham George Smith, Christian Giese, Dorte Bodin Dresbøll, Niels Alvin Faircloth Olsen, Miriam Athmann |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2022 |
| Předmět: |
Buildings and machinery
phenotyping root nodule Physiology Computer science Plant Science rhizotron Machine Learning Annotation Soil Deep Learning GUI Image Processing Computer-Assisted Farm nutrient management Segmentation biopore business.industry interactive machine learning Deep learning segmentation deep learning Pattern recognition Image segmentation Breeding and genetics Neural Networks Computer Artificial intelligence business |
| Zdroj: | Smith, A G, Han, E, Petersen, J, Olsen, N A F, Giese, C, Athmann, M, Dresbøll, D B & Thorup-Kristensen, K 2022, ' RootPainter : deep learning segmentation of biological images with corrective annotation ', New Phytologist, vol. 236, pp. 774-791 . https://doi.org/10.1111/nph.18387 New Phytologist |
| DOI: | 10.1111/nph.18387 |
| Popis: | Convolutional neural networks (CNNs) are a powerful tool for plant image analysis, but challenges remain in making them more accessible to researchers without a machine-learning background. We present RootPainter, an open-source graphical user interface based software tool for the rapid training of deep neural networks for use in biological image analysis. We evaluate RootPainter by training models for root length extraction from chicory (Cichorium intybus L.) roots in soil, biopore counting, and root nodule counting. We also compare dense annotations with corrective ones that are added during the training process based on the weaknesses of the current model. Five out of six times the models trained using RootPainter with corrective annotations created within 2 h produced measurements strongly correlating with manual measurements. Model accuracy had a significant correlation with annotation duration, indicating further improvements could be obtained with extended annotation. Our results show that a deep-learning model can be trained to a high accuracy for the three respective datasets of varying target objects, background, and image quality with < 2 h of annotation time. They indicate that, when using RootPainter, for many datasets it is possible to annotate, train, and complete data processing within 1 d. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|