Результаты исследований: Научные публикации в периодических изданиях › Обзорная статья › Рецензирование
Plant Phenomics: Fundamental Bases, Software and Hardware Platforms, and Machine Learning. / Demidchik, V. V.; Shashko, A. Y.; Bandarenka, U. Y.; Smolikova, G. N.; Przhevalskaya, D. A.; Charnysh, M. A.; Pozhvanov, G. A.; Barkosvkyi, A. V.; Smolich, I. I.; Sokolik, A. I.; Yu, M.; Medvedev, S. S.
в: Russian Journal of Plant Physiology, Том 67, № 3, 05.2020, стр. 397-412.Результаты исследований: Научные публикации в периодических изданиях › Обзорная статья › Рецензирование
}
TY - JOUR
T1 - Plant Phenomics: Fundamental Bases, Software and Hardware Platforms, and Machine Learning
AU - Demidchik, V. V.
AU - Shashko, A. Y.
AU - Bandarenka, U. Y.
AU - Smolikova, G. N.
AU - Przhevalskaya, D. A.
AU - Charnysh, M. A.
AU - Pozhvanov, G. A.
AU - Barkosvkyi, A. V.
AU - Smolich, I. I.
AU - Sokolik, A. I.
AU - Yu, M.
AU - Medvedev, S. S.
N1 - Demidchik, V.V., Shashko, A.Y., Bandarenka, U.Y. et al. Plant Phenomics: Fundamental Bases, Software and Hardware Platforms, and Machine Learning. Russ J Plant Physiol 67, 397–412 (2020). https://doi.org/10.1134/S1021443720030061
PY - 2020/5
Y1 - 2020/5
N2 - In recent years, a new branch of plant physiology, plant phenomics, which focuses on identifying patterns of organization and changes in plant Phenomes, i.e., physical and biochemical characteristics, considered as a set of phenotypes of a plant organism, has emerged. Phenomics is a postgenomic discipline that actively uses the achievements of the genomic era and bioinformatics. It supplements them with standardized and statistically significant factual material on phenotypes with a high degree of detail. The technique of obtaining and analyzing information about phenotypes in phenomics is called phenotyping. High-performance phenotyping, providing digital automated analysis of large data samples, has become widespread. Recent progress in high-performance phenotyping has been associated with the development of image registration systems in various spectral regions, approaches to cultivating plant objects under standardized conditions, sensory technologies, robotics, and methods for data processing and analysis, such as computer vision and machine learning (artificial neural network). Phenomics technologies have a high information content analysis, surpassing human capabilities, performing measurements in the hyperspectral range using X-ray tomography and ultra-precise "thermal" images, and have a number of other low-invasive and precision approaches. Arrays of data obtained using phenomics technologies are recorded and processed automatically and are free from the problems of subjective assessment and inadequate statistical processing. It is assumed that phenotyping will allow for the creation of digital models of the vital activity processes and the "formation" of plant productivity at the organism level in connection with the dynamics of transcriptomes, proteomes, and metabolomes. Phenomics helps researchers transform a large amount of information received from modern sensors into new knowledge using computer data processing and modeling, reducing the distance from basic science to the practical application of results in crop production and breeding. Phenotyping is actively developing both in laboratory and in greenhouse conditions as well as on open agricultural sites, forests, and in real natural phytocenoses. The review analyzes the current state of plant phenomics with a focus on technical aspects, in particular, the design of hardware-software phenotyping complexes, i.e., phenomics platforms, as well as the use of neural networks in phenotyping of plant organisms.
AB - In recent years, a new branch of plant physiology, plant phenomics, which focuses on identifying patterns of organization and changes in plant Phenomes, i.e., physical and biochemical characteristics, considered as a set of phenotypes of a plant organism, has emerged. Phenomics is a postgenomic discipline that actively uses the achievements of the genomic era and bioinformatics. It supplements them with standardized and statistically significant factual material on phenotypes with a high degree of detail. The technique of obtaining and analyzing information about phenotypes in phenomics is called phenotyping. High-performance phenotyping, providing digital automated analysis of large data samples, has become widespread. Recent progress in high-performance phenotyping has been associated with the development of image registration systems in various spectral regions, approaches to cultivating plant objects under standardized conditions, sensory technologies, robotics, and methods for data processing and analysis, such as computer vision and machine learning (artificial neural network). Phenomics technologies have a high information content analysis, surpassing human capabilities, performing measurements in the hyperspectral range using X-ray tomography and ultra-precise "thermal" images, and have a number of other low-invasive and precision approaches. Arrays of data obtained using phenomics technologies are recorded and processed automatically and are free from the problems of subjective assessment and inadequate statistical processing. It is assumed that phenotyping will allow for the creation of digital models of the vital activity processes and the "formation" of plant productivity at the organism level in connection with the dynamics of transcriptomes, proteomes, and metabolomes. Phenomics helps researchers transform a large amount of information received from modern sensors into new knowledge using computer data processing and modeling, reducing the distance from basic science to the practical application of results in crop production and breeding. Phenotyping is actively developing both in laboratory and in greenhouse conditions as well as on open agricultural sites, forests, and in real natural phytocenoses. The review analyzes the current state of plant phenomics with a focus on technical aspects, in particular, the design of hardware-software phenotyping complexes, i.e., phenomics platforms, as well as the use of neural networks in phenotyping of plant organisms.
KW - computer vision
KW - convolutional neural networks
KW - high-performance phenotyping
KW - machine learning
KW - phenomics platform
KW - plant phenomics
KW - plants
KW - CHLOROPHYLL FLUORESCENCE
KW - DROUGHT STRESS
KW - QUANTIFICATION
KW - CLASSIFICATION
KW - RESPONSES
KW - SHAPE
KW - WATER-STRESS
KW - GROWTH
KW - ARCHITECTURE
KW - LIGHT ACCLIMATION
UR - http://www.scopus.com/inward/record.url?scp=85084844241&partnerID=8YFLogxK
UR - https://www.mendeley.com/catalogue/980167a1-29fb-3fcd-ad82-bd2c829639a0/
U2 - 10.1134/S1021443720030061
DO - 10.1134/S1021443720030061
M3 - Обзорная статья
VL - 67
SP - 397
EP - 412
JO - Russian Journal of Plant Physiology
JF - Russian Journal of Plant Physiology
SN - 1021-4437
IS - 3
ER -
ID: 53520288