Automated Machine Learning for High-Throughput Image-Based Plant Phenotyping

Abstract

Automated machine learning (AutoML) has been heralded as the next wave in artificial intelligence with its promise to deliver high-performance end-to-end machine learning pipelines with minimal effort from the user. However, despite AutoML showing great promise for computer vision tasks, to the best of our knowledge, no study has used AutoML for image-based plant phenotyping. To address this gap in knowledge, we examined the application of AutoML for image-based plant phenotyping using wheat lodging assessment with unmanned aerial vehicle (UAV) imagery as an example. The performance of an open-source AutoML framework, AutoKeras, in image classification and regression tasks was compared to transfer learning using modern convolutional neural network (CNN) architectures. For image classification, which classified plot images as lodged or non-lodged, transfer learning with Xception and DenseNet-201 achieved the best classification accuracy of 93.2%, whereas AutoKeras had a 92.4% accuracy. For image regression, which predicted lodging scores from plot images, transfer learning with DenseNet-201 had the best performance (R2 = 0.8303, root mean-squared error (RMSE) = 9.55, mean absolute error (MAE) = 7.03, mean absolute percentage error (MAPE) = 12.54%), followed closely by AutoKeras (R2 = 0.8273, RMSE = 10.65, MAE = 8.24, MAPE = 13.87%). In both tasks, AutoKeras models had up to 40-fold faster inference times compared to the pretrained CNNs. AutoML has significant potential to enhance plant phenotyping capabilities applicable in crop breeding and precision agriculture.

AmazonCRIME: un conjunto de datos y punto de referencia de Inteligencia Artificial Geoespacial para la clasificación de áreas potenciales vinculadas a Crímenes Ambientales Transnacionales en la Selva Amazónica

“…La arquitectura DenseNet-201 (Huang et al, 2017), se seleccionó para el entrenamiento de los modelos, debido a sus capacidades para mejorar la precisión en tareas de clasificación de imágenes. Varios trabajos (Abdani et al, 2019;Koh et al, 2021;Khan et al, 2021) han demostrado que las CNNs pueden ser esencialmente más profundas, más precisas y eficientes de entrenar si contienen conexiones más cortas entre las capas cercanas a la entrada y las cercanas a la salida, este es el principio de la arquitectura DenseNet-201, conectar cada capa con todas las demás en forma de retroalimentación. Entre sus principales ventajas están el hecho de que este tipo de red reduce el problema de la desaparición de los gradientes, fortalece la propagación de características, promueve la reutilización de funciones y reduce considerablemente el número de parámetros de entrenamiento (Huang et al, 2017).…”

Deep Neural Networks to Detect Weeds from Crops in Agricultural Environments in Real-Time: A Review

“…The model was originally developed to classify medical images, but is also used in the agricultural field due to its excellent efficiency. Koh et al [119] successfully used this model for high-throughput image-based plant phenotyping. Shah et al [120] described AmoebaNetalso, which refers to algorithms used for the automatic creation of neural networks.…”

Artificial Intelligence Meets Marine Ecotoxicology: Applying Deep Learning to Bio-Optical Data from Marine Diatoms Exposed to Legacy and Emerging Contaminants

“…These optical biomarkers have been used from field to experimental trials using a wide array of organisms from microalgae to higher plants [ 9 , 10 , 14 ]. Conventional data analysis pipelines typically involve computer vision tasks (e.g., wheat head counting using object detection), which are addressed through the development of signal processing or machine learning (ML) algorithms [ 19 ]. When exposed to anthropogenic contaminants the typical chlorophyll a induction curves suffer alterations, not only on their intensity values but also on their shape [ 9 ].…”