Rectified Meta-Learning from Noisy Labels for Robust Image-based Plant Disease Diagnosis

TL;DR

This paper introduces a rectified meta-learning framework integrated into CNNs to enhance robustness against noisy labels in plant disease image classification, improving accuracy without extra supervision.

Contribution

It proposes a novel rectified meta-learning module that focuses on unbiased samples, is assumption-free regarding label noise, and is easily integrable into existing deep models.

Findings

Outperforms state-of-the-art methods in noisy label scenarios.

Accelerates convergence and improves classification accuracy.

Works effectively across various noise types.

Abstract

Plant diseases serve as one of main threats to food security and crop production. It is thus valuable to exploit recent advances of artificial intelligence to assist plant disease diagnosis. One popular approach is to transform this problem as a leaf image classification task, which can be then addressed by the powerful convolutional neural networks (CNNs). However, the performance of CNN-based classification approach depends on a large amount of high-quality manually labeled training data, which are inevitably introduced noise on labels in practice, leading to model overfitting and performance degradation. To overcome this problem, we propose a novel framework that incorporates rectified meta-learning module into common CNN paradigm to train a noise-robust deep network without using extra supervision information. The proposed method enjoys the following merits: i) A rectified meta-learning is designed to pay more attention to unbiased samples, leading to accelerated convergence and improved classification accuracy. ii) Our method is free on assumption of label noise distribution, which works well on various kinds of noise. iii) Our method serves as a plug-and-play module, which can be embedded into any deep models optimized by gradient descent based method. Extensive experiments are conducted to demonstrate the superior performance of our algorithm over the state-of-the-arts.