A deep learning approach for detection and localization of leaf anomalies

TL;DR

This paper explores unsupervised deep learning models, specifically autoencoders, for detecting and localizing leaf diseases in crops, demonstrating that vector-quantized variational autoencoders outperform other models on a public dataset.

Contribution

It introduces an unsupervised approach using three autoencoder types for leaf anomaly detection and localization, highlighting the superior performance of VQ-VAE.

Findings

VQ-VAE outperforms CAE and CVAE in detection and localization

Unsupervised models can effectively identify leaf anomalies without labeled data

The approach is validated on a public dataset of pepper and cherry leaves

Abstract

The detection and localization of possible diseases in crops are usually automated by resorting to supervised deep learning approaches. In this work, we tackle these goals with unsupervised models, by applying three different types of autoencoders to a specific open-source dataset of healthy and unhealthy pepper and cherry leaf images. CAE, CVAE and VQ-VAE autoencoders are deployed to screen unlabeled images of such a dataset, and compared in terms of image reconstruction, anomaly removal, detection and localization. The vector-quantized variational architecture turns out to be the best performing one with respect to all these targets.