n of experiments. Section 4 analyzes all research findings, proposes possible improvements, and suggests future research directions. Finally, Section 5 summarizes the main contributions of this study. 2 Materials and methods 2.1 Datasets The chili leaf disease dataset is sourced from publicly available datasets on the internet( https://www.kaggle.com/datasets/dhenyd/chili-plant-disease/ ). The initial dataset consists of 500 images, encompassing five categories: healthy chili leaves, leaf curl disease, leaf spot disease, whitefly, and yellowing disease. Samples of each category are illustrated in Figure 1 . Figure 1 Chili samples. (A) Healthy chili leaves. (B) Leaf curl disease. (C) Leaf spo
Open resource ↗Kaggle · dhenyd/chili-plant-disease · lines:59-78Unverified paper record
MCCM: multi-scale feature extraction network for disease classification and recognition of chili leaves.
Frontiers in plant science · 28 May 2024 · 10.3389/fpls.2024.1367738
Abstract
Currently, foliar diseases of chili have significantly impacted both yield and quality. Despite effective advancements in deep learning techniques for the classification of chili leaf diseases, most existing classification models still face challenges in terms of accuracy and practical application in disease identification. Therefore, in this study, an optimized and enhanced convolutional neural network model named MCCM (MCSAM-ConvNeXt-MSFFM) is proposed by introducing ConvNeXt. The model incorporates a Multi-Scale Feature Fusion Module (MSFFM) aimed at better capturing disease features of various sizes and positions within the images. Moreover, adjustments are made to the positioning, activation functions, and normalization operations of the MSFFM module to further optimize the overall model. Additionally, a proposed Mixed Channel Spatial Attention Mechanism (MCSAM) strengthens the correlation between non-local channels and spatial features, enhancing the model's extraction of fundamental characteristics of chili leaf diseases. During the training process, pre-trained weights are obtained from the Plant Village dataset using transfer learning to accelerate the model's convergence. Regarding model evaluation, the MCCM model is compared with existing CNN models (Vgg16, ResNet34, GoogLeNet, MobileNetV2, ShuffleNet, EfficientNetV2, ConvNeXt), and Swin-Transformer. The results demonstrate that the MCCM model achieves average improvements of 3.38%, 2.62%, 2.48%, and 2.53% in accuracy, precision, recall, and F1 score, respectively. Particularly noteworthy is that compared to the original ConvNeXt model, the MCCM model exhibits significant enhancements across all performance metrics. Furthermore, classification experiments conducted on rice and maize disease datasets showcase the MCCM model's strong generalization performance. Finally, in terms of application, a chili leaf disease classification website is successfully developed using the Flask framework. This website accurately identifies uploaded chili leaf disease images, demonstrating the practical utility of the model.
Plant phenotyping relevance
トウガラシ葉の病徴を画像から分類する深層学習手法を開発・比較評価しており、植物の病害状態を推定するフェノタイピング手法が中心である。
abstractan optimized and enhanced convolutional neural network model named MCCM (MCSAM-ConvNeXt-MSFFM) is proposed
abstractThe model incorporates a Multi-Scale Feature Fusion Module (MSFFM) aimed at better capturing disease features of various sizes and positions within the images.
abstractRegarding model evaluation, the MCCM model is compared with existing CNN models
abstracta chili leaf disease classification website is successfully developed using the Flask framework
Code and data availability
The paper's phenotyping inputs are public Kaggle image datasets: the chili leaf disease dataset used as the primary experimental data, plus rice and maize leaf disease datasets used for generalization tests. No author analysis code, trained model checkpoints, or dataset deposits are stated; the Flask website is a live-
the corresponding accuracy rate of each cross-validation. 3.10.2 Rice and maize disease test results To validate the generalization ability of the MCCM model on different crop diseases, this study conducted performance tests using publicly available datasets, including rice leaf disease dataset and maize leaf disease dataset ( https://www.kaggle.com/datasets/nirmalsankalana/rice-leaf-disease-image and https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset ). The rice leaf disease dataset covers four types: rice blast, bacterial leaf blight, brown spot, and rice tungro disease, while the maize leaf disease dataset includes healthy states and three types of diseases:
Open resource ↗Kaggle · nirmalsankalana/rice-leaf-disease-image · lines:456-477aize disease test results To validate the generalization ability of the MCCM model on different crop diseases, this study conducted performance tests using publicly available datasets, including rice leaf disease dataset and maize leaf disease dataset ( https://www.kaggle.com/datasets/nirmalsankalana/rice-leaf-disease-image and https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset ). The rice leaf disease dataset covers four types: rice blast, bacterial leaf blight, brown spot, and rice tungro disease, while the maize leaf disease dataset includes healthy states and three types of diseases: gray leaf spot, rust, and leaf blight. To ensure dataset diversity, data a
Open resource ↗Kaggle · smaranjitghose/corn-or-maize-leaf-disease-dataset · lines:456-477This is an automatically classified, unverified record. Curator approval is required before any resource enters the Catalog.