Analysis of Learned Features and Framework for Potato Disease Detection

TL;DR

This paper compares RCNN and attention-based networks for potato disease detection, focusing on feature learning to handle dataset shifts, achieving high accuracy on both training and unseen data.

Contribution

It introduces a framework using RCNN and attention networks to improve feature learning for potato disease detection under dataset shift conditions.

Findings

Achieved approximately 95% accuracy on test data similar to training data.

Maintained 84% accuracy on unseen dataset, demonstrating robustness.

Both classifiers performed comparably across different datasets.

Abstract

For applications like plant disease detection, usually, a model is trained on publicly available data and tested on field data. This means that the test data distribution is not the same as the training data distribution, which affects the classifier performance adversely. We handle this dataset shift by ensuring that the features are learned from disease spots in the leaf or healthy regions, as applicable. This is achieved using a faster Region-based convolutional neural network (RCNN) as one of the solutions and an attention-based network as the other. The average classification accuracies of these classifiers are approximately 95% while evaluated on the test set corresponding to their training dataset. These classifiers also performed equivalently, with an average score of 84% on a dataset not seen during the training phase.