Interactive machine learning for soybean seed and seedling quality classification

New computer vision solutions combined with artificial intelligence algorithms can help recognize patterns in biological images, reducing subjectivity and optimizing the analysis process. The aim of this study was to propose an approach based on interactive and traditional machine learning methods to classify soybean seeds and seedlings according to their appearance and physiological potential. In addition, we correlated the appearance of seeds to their physiological performance. Images of soybean seeds and seedlings were used to develop models using low-cost approaches and free-access software. The models developed showed high performance, with overall accuracy reaching 0.94 for seeds and seedling classification. The high precision of the models that were developed based on interactive and traditional machine learning demonstrated that the method can easily be used to classify soybean seeds according to their appearance, as well as to classify soybean seedling vigor quickly and non-subjectively. The appearance of soybean seeds is strongly correlated with their physiological performance.

[1]  Naoshi Kondo,et al.  Machine vision based soybean quality evaluation , 2017, Comput. Electron. Agric..

[2]  Isabella G. S. Visschers,et al.  An objective high‐throughput screening method for thrips damage quantitation using Ilastik and ImageJ , 2018 .

[3]  Martin Horn,et al.  Integration of the ImageJ Ecosystem in KNIME Analytics Platform , 2020, Frontiers in Computer Science.

[4]  Santosh Shrestha,et al.  Multispectral imaging – a new tool in seed quality assessment? , 2018, Seed Science Research.

[5]  Márcio Lúcio Dias Pereira,et al.  SAPL®: um software gratuito para determinação do potencial fisiológico em sementes de soja , 2018 .

[6]  Pierre Geurts,et al.  Supervised learning with decision tree-based methods in computational and systems biology. , 2009, Molecular bioSystems.

[7]  Andreas Holzinger,et al.  Interactive machine learning: experimental evidence for the human in the algorithmic loop , 2018, Applied Intelligence.

[8]  Satrajit S. Ghosh,et al.  Everything Matters: The ReproNim Perspective on Reproducible Neuroimaging , 2018, Front. Neuroinform..

[9]  Moon S. Kim,et al.  Prediction of crude protein and oil content of soybeans using Raman spectroscopy , 2013 .

[10]  C. Klukas,et al.  Advanced phenotyping and phenotype data analysis for the study of plant growth and development , 2015, Front. Plant Sci..

[11]  K. Asefpour Vakilian.  Machine learning improves our knowledge about miRNA functions towards plant abiotic stresses , 2020, Scientific Reports.

[12]  M. Kim,et al.  Non-destructive technique for determining the viability of soybean (Glycine max) seeds using FT-NIR spectroscopy. , 2018, Journal of the science of food and agriculture.

[13]  Jan G. Bjaalie,et al.  QUINT: Workflow for Quantification and Spatial Analysis of Features in Histological Images From Rodent Brain , 2019, Front. Neuroinform..

[14]  F. G. Gomes-Júnior,et al.  Vigor-S, a new system for evaluating the physiological potential of maize seeds , 2018 .

[15]  Yordan I Yordanov,et al.  Hep G2 cell culture confluence measurement in phase-contrast micrographs – a user-friendly, open-source software-based approach , 2019, Toxicology mechanisms and methods.

[16]  André Dantas de Medeiros,et al.  SAPL®: a free software for determining the physiological potential in soybean seeds1 , 2018, Pesquisa Agropecuária Tropical.

[17]  Aoife A Gowen,et al.  Feasibility of conventional and Roundup Ready® soybeans discrimination by different near infrared reflectance technologies. , 2012, Food chemistry.

[18]  Robert L. Geneve,et al.  Near-infrared spectroscopy used to predict soybean seed germination and vigour , 2018, Seed Science Research.

[19]  Aldenor G. Santos,et al.  Occurrence of the potent mutagens 2- nitrobenzanthrone and 3-nitrobenzanthrone in fine airborne particles , 2019, Scientific Reports.

[20]  Qiugang Lu,et al.  Support vector machine approach for model-plant mismatch detection , 2020, Comput. Chem. Eng..

[21]  K. Fujimura,et al.  A system for automated seed vigour assessment , 2001 .

[22]  D. Dias,et al.  Quality classification of Jatropha curcas seeds using radiographic images and machine learning , 2020 .

[23]  I. D. Menezes,et al.  Effect of tear/crack on soybean (Glycine max) seed coat, physiological quality and pathology of the seed , 2019, June 2019.

[24]  Boyd Blackwell,et al.  Observations on water distribution in soybean seed during hydration processes using nuclear magnetic resonance imaging , 2002 .

[25]  Andreas Holzinger,et al.  Interactive machine learning for health informatics: when do we need the human-in-the-loop? , 2016, Brain Informatics.

[26]  Liang Tong,et al.  Identification of Seed Maize Fields With High Spatial Resolution and Multiple Spectral Remote Sensing Using Random Forest Classifier , 2020, Remote. Sens..

[27]  Shveta Mahajan,et al.  Machine vision based alternative testing approach for physical purity, viability and vigour testing of soybean seeds (Glycine max) , 2018, Journal of Food Science and Technology.

[28]  Li Xiaoli,et al.  Rapidly and exactly determining postharvest dry soybean seed quality based on machine vision technology , 2019, Scientific Reports.

[29]  Ming Chen,et al.  Data-mining Techniques for Image-based Plant Phenotypic Traits Identification and Classification , 2019, Scientific Reports.

[30]  M. Maheswari,et al.  Characterisation of Soybean and Wheat Seeds by Nuclear Magnetic Resonance Spectroscopy , 2004, Biologia Plantarum.

[31]  E. Finch-SavageW..  Seed vigour and crop establishment – extending performance beyond adaptation , 2022 .

[32]  A. Caverzan,et al.  Physiologic alterations in orthodox seeds due to deterioration processes. , 2019, Plant physiology and biochemistry : PPB.

[33]  Rohit Bhargava,et al.  Protein and oil composition predictions of single soybeans by transmission Raman spectroscopy. , 2012, Journal of agricultural and food chemistry.

[34]  Lianxing Gao,et al.  Discriminating and elimination of damaged soybean seeds based on image characteristics , 2015 .

[35]  Fred A. Hamprecht,et al.  ilastik: interactive machine learning for (bio)image analysis , 2019, Nature Methods.