PLANT PHENOTYPING BY AN ADAPTIVE IMAGE PROCESSING SYSTEM BASED ON CONVOLUTIONAL NEURAL NETWORKS
https://doi.org/10.34822/1999-7604-2021-2-6-16
Abstract
The article continues the study devoted to the design of automatic plant phenotyping systems. This topic of scientific research is now widely developed. The intensive development of digital technologies in agribusiness can significantly increase labor productivity, reduce errors in the visual analysis of plant health, and as a result, increase productivity. It is commonly known that building an automatic plant phenotyping system requires mass analysis of a large number (about a thousand) of plants through digital image processing. As a key element of the automatic phenotyping system, the convolutional neural network apparatus is used. This apparatus analyzes the characteristics of plant leaves under the influence of the external environment for the presence of diseases and morphological features of the plant. The analysis of the quality of the automatic phenotyping system based on convolutional neural networks shows a high efficiency of plant disease assessment. As a component of the adaptation of the control system, we use the E-network graph, which allows forming a control vector for the technical vision system and the development of agronomic recommendations.
About the Authors
M. Ya. Braginsky
Surgut State University, Surgut
Russian Federation
Russian Federation
E-mail: mick17@mail.ru
D. V. Tarakanov
Surgut State University, Surgut
Russian Federation
Russian Federation
References
1. Афонников Д. А., Генаев М. А., Дорошков А. В., Комышев Е. Г., Пшеничникова Т. А. Методы высокопроизводительного фенотипирования растений для массовых селекционногенетических экспериментов // Генетика. 2016. Т. 52, № 7. C. 788–803.
2. Ashok R., Uma S. K. Garden Environmental Monitoring & Automatic Control System Using Sensors // International Journal for Modern Trends in Science and Technology (IJMTST). 2016. Vol. 2, № 5. Р. 141–144.
3. Faouzi D., Bibi-Triki N., Draoui B., Abène A. Greenhouse Environmental Control Using Optimized, Modeled and Simulated Fuzzy Logic Controller Technique in MATLAB SIMULINK // Computer Technology and Application. 2016. № 7. Р. 273–286.
4. Запевалова Л. Ю., Запевалов А. В. Предварительная обработка цифровых изображений в системе управления вертикальной фермой // Вестник кибернетики. 2020. № 1. C. 77–84.
5. Setiawan W., Utoyo M. I., Rulaningtyas R. Reconfiguration Layers of Convolutional Neural Network for Fundus Patches Classification // Bulletin of Electrical Engineering and Informatics. 2021. № 10 (1). Р. 383–389.
6. SqueezeNet-Residual by songhan. URL: http://songhan.github.io/SqueezeNet-Residual (дата обращения: 06.05.2021).
7. Брагинский М. Я., Тараканов Д. В. Концепция построения аналитико-имитационных моделей человеко-машинных систем управления с помощью Е-сетей // Вестник кибернетики. 2020. № 2. C. 50–57.
8. Gilbert D., Heiner M., Ghanbar L., Chodak J. Spatial Quorum Sensing Modelling using Coloured Hybrid Petri Nets and Simulative Model Checking // BMC Bioinformatics. 2019. № 20 (4). P. 173.
9. Брагинский М. Я., Тараканов Д. В. Оценка состояния растений с использованием сверточных нейронных сетей // Вестник кибернетики. 2021. № 1. С. 41–50. DOI 10.34822/ 1999-7604-2021-1-41-50.
10. Ronneberger O., Fischer P., Brox T. U-Net: Convolutional Networks for Biomedical Image Segmentation. URL: https://arxiv.org/abs/1505.04597 (дата обращения: 06.05.2021).
11. Брагинский М. Я., Тараканов Д. В. Компьютерная обработка изображений в задаче выявления фитопатологии биологических культур // Интеграция науки, образования, общества, производства и экономики : сб. ст. по материалам междунар. науч.-практич. конф. Уфа, 2020. С. 29–35.
Review
For citations:
Braginsky M.Ya., Tarakanov D.V. PLANT PHENOTYPING BY AN ADAPTIVE IMAGE PROCESSING SYSTEM BASED ON CONVOLUTIONAL NEURAL NETWORKS. Proceedings in Cybernetics. 2021;(2 (42)):6-16. (In Russ.) https://doi.org/10.34822/1999-7604-2021-2-6-16
Views:
374
Email this article 