Crop Lodging Prediction From UAV-Acquired Images of Wheat and Canola Using a DCNN Augmented With Handcrafted Texture Features

Abstract

Lodging, the permanent bending over of food crops, leads to poor plant growth and development. Consequently, lodging results in reduced crop quality, lowers crop yield, and makes harvesting difficult. Plant breeders routinely evaluate several thousand breeding lines, and therefore, automatic lodging detection and prediction is of great value aid in selection. In this paper, we propose a deep convolutional neural network (DCNN) architecture for lodging classification using five spectral channel orthomosaic images from canola and wheat breeding trials. Also, using transfer learning, we trained 10 lodging detection models using well-established deep convolutional neural network architectures. Our proposed model outperforms the stateof-the-art lodging detection methods in the literature that use only handcrafted features. In comparison to 10 DCNN lodging detection models, our proposed model achieves comparable results while having a substantially lower number of parameters. This makes the proposed model suitable for applications such as real-time classification using inexpensive hardware for high-throughput phenotyping pipelines. The GitHub repository at https://github. com/FarhadMaleki/LodgedNet contains code and models.Co-first authors.

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

“…allowing variations in color to potentially contribute more informative features for modelling [20,26]. Results in our study suggest that NAS generated models (AutoML) can provide very competitive performance compared to modern human designed CNN architectures.…”

“…showed that the AutoKeras model had a much higher number of predictions exceeding the maximum value of 100 (n=39), with the largest predicted value being 118 whereas the ResNet-50 model only have one prediction exceeding 100 with a value of 100.29 ( Figure 10). This would likely explain why the MAE score, which is the average of the absolute difference between predicted scores and actual learning [26]. The higher classification accuracies reported in that study may be due to the use of image data augmentation (i.e.…”

“…Although transfer learning using existing modern CNN architectures provides a good starting point in computer vision tasks, a major limitation is that these CNN models have been trained using 3channels RGB images and this prevents direct application of these models for image sources beyond 3-channels, such as multispectral and hyperspectral images [26]. In addition, existing CNN architectures may not always provide the best performance compared to custom-designed models.…”

Breast Cancer Diagnosis with Transfer Learning and Global Pooling

“…The high morbidity and considerable healthcare cost associated with cancer has inspired researchers to develop more accurate models for cancer detection. Over the last five years, data mining and machine learning models have been used in a variety of research areas to dramatically improve our ability to discover emergent patterns within large datasets [3]- [8]. Developing computer-aided diagnosis (CAD) systems, integrated with medical image computing and machine learning methods, has become one of the major research paradigms for life-critical diagnosis [9].…”

A Hybrid Deep Learning Architecture for Leukemic B-lymphoblast Classification

“…Therefore, developing accurate and reliable approaches for Leukemia detection is important for early treatment. Numerous study results showed that with the advancement of computational capabilities, hidden trends, patterns and relationships can be discovered using the application of data mining approaches in many different areas [5]- [8]. described in the subsequent sections.…”