Seed-per-pod estimation for plant breeding using deep learning

• Published in: Computers and electronics in agriculture 2018-07, Vol.150, p.196-204
• Main Authors: Uzal, L.C., Grinblat, G.L., Namías, R., Larese, M.G., Bianchi, J.S., Morandi, E.N., Granitto, P.M.
• Format: Article
• Language: English
• Subjects: Accuracy; Artificial neural networks; Computer vision; Deep learning; Estimates; Feature extraction; Flowers & plants; Genotype & phenotype; Machine vision; Neural networks; Plant breeding programs; Plant phenotyping; Plant reproduction; Seeds; Soybean; Soybeans; Support vector machines
• ISSN: 0168-1699; 1872-7107
• DOI: 10.1016/j.compag.2018.04.024

[Display omitted] •Deep learning method for estimating the number of seeds into soybean pods.•The deep learning approach has improved generalization capability.•Apart from network depth, other hyperparameters selection is not critical.•Responses of the network were interpreted with a simple visualization technique.•Our results suggest that deep learning is a promising technique in plant phenotyping. Commercial and scientific plant breeding programs require the phenotyping of large populations. Phenotyping is typically a manual task (costly, time-consuming and sometimes arbitrary). The use of computer vision techniques is a potential solution to some of these specific tasks. In the last years, Deep Learning, and in particular Convolutional Neural Networks (CNNs), have shown a number of advantages over traditional methods in the area. In this work we introduce a computer vision method that estimates the number of seeds into soybean pods, a difficult task that usually requires the intervention of human experts. To this end we developed a classic approach, based on tailored features extraction (FE) followed by a Support Vector Machines (SVM) classification model, and also the referred CNNs. We show how standard CNNs can be easily configured and how a simple method can be used to visualize the key features learned by the model in order to infer the correct class. We processed different seasons batches with both methods obtaining 50.4% (FE + SVM) and 86.2% (CNN) of accuracy in test, highlighting the particularly high increase in generalization capabilities of a deep learning approach over a classic machine vision approach in this task. Dataset and code are publicly available.