Crop Disease Recognition Based on Modified Light-Weight CNN With Attention Mechanism

Yang Liu,Guoqin Gao,Zhenhui Zhang

DOI: https://doi.org/10.1109/access.2022.3216285

IF: 3.9

2022-11-04

IEEE Access

Abstract:The agricultural production is greatly affected by various plant diseases. Classifying the severity of crop diseases is the requirement for formulating disease prevention and control strategies. However, the differences between different severity of the same crop disease are very tiny. It increases the difficulty of correct crop disease recognition. For example, at the early stage of the disease, the lesions on the leaves are not obvious. And it is very difficult to extract the features of the lesions. However, these very small color and texture differences of the lesions are the key patterns to distinguish different kinds of diseases of the same species. In order to achieve better performance in the fine-grained classification of the crop diseases, a modified light-weight convolution neural network was proposed. Multi-scale convolution kernel and coordinate attention mechanism are introduced in SqueezeNext to extract the features of the lesions accurately. The performance of the proposed model was evaluated using the AI challenger 2018 plant disease recognition dataset, and the recognition accuracy can reach 91.94%, which is 3.02% point higher than the original SqueezeNext model. In order to verify the effectiveness of the proposed model, comparative experiments were carried out using ReseNet50, Xception and mobilenetv2. The experimental results showed that the accuracy of the proposed method was slightly better than Xception, while the model size is only 2.83 MB, which is only 3.45% of Xception. The proposed method balances the performance and efficiency very well. Thus, it is suitable for deployment on mobile terminals and other embedded resource-constrained devices, which help to promote the popularization of smart agriculture application.

computer science, information systems,telecommunications,engineering, electrical & electronic

What problem does this paper attempt to address?

The problem this paper attempts to address is the fine-grained classification challenge in crop disease identification. Specifically, the differences between the same crop disease at different severity levels are very subtle, which increases the difficulty of correctly identifying crop diseases. For example, in the early stages of a disease, the lesions on the leaves are not obvious, making it difficult to extract features of these lesions. However, these very small color and texture differences are key patterns for distinguishing different diseases of the same species. To achieve better performance in fine-grained classification, the authors propose an improved lightweight Convolutional Neural Network (CNN) that accurately extracts lesion features by introducing multi-scale convolutional kernels and a coordinate attention mechanism. The main contributions of the paper include: 1. **Improved lightweight CNN model**: An improved model based on SqueezeNext is proposed, introducing multi-scale convolutional kernels and a coordinate attention mechanism. 2. **High accuracy and low computational complexity**: On the AI Challenger 2018 plant disease recognition dataset, the model achieved an identification accuracy of 91.94%, which is 3.02 percentage points higher than the original SqueezeNext model, while the model size is only 2.83 MB, making it suitable for deployment on mobile terminals and other embedded resource-constrained devices. 3. **Balance of effectiveness and efficiency**: Comparative experiments with models such as ResNet50, Xception, and MobileNetV2 validate the effectiveness of the proposed method. Experimental results show that the method maintains high accuracy while having lower computational complexity and smaller model size. In summary, this paper aims to improve the accuracy of crop disease identification, especially in fine-grained classification tasks, through an improved lightweight CNN model, while maintaining low computational complexity and model size to meet the needs of practical applications.

Crop Disease Recognition Based on Modified Light-Weight CNN With Attention Mechanism

Crop Disease Identification by Fusing Multiscale Convolution and Vision Transformer.

HLNet Model and Application in Crop Leaf Diseases Identification

Multi-Plant Disease Identification Based on Lightweight ResNet18 Model

Crop leaf disease grade identification based on an improved convolutional neural network

Lightweight Inception Networks for the Recognition and Detection of Rice Plant Diseases

Development of a Lightweight Crop Disease Image Identification Model Based on Attentional Feature Fusion

Improved EfficientNet for corn disease identification

A lightweight dual-attention network for tomato leaf disease identification

Improved MobileNetV2 crop disease identification model for intelligent agriculture

DFCANet: A Novel Lightweight Convolutional Neural Network Model for Corn Disease Identification

Corn Plant Disease Recognition Based on Migration Learning and Convolutional Neural Network

Plant Foliage Disease Diagnosis Using Light-Weight Efficient Sequential CNN Model

Crop Organ Segmentation and Disease Identification Based on Weakly Supervised Deep Neural Network

A Lightweight Attention-Based Convolutional Neural Networks for Tomato Leaf Disease Classification

Enhancing practicality of deep learning for crop disease identification under field conditions: insights from model evaluation and crop-specific approaches

An Industrial-Grade Solution for Crop Disease Image Detection Tasks

Classification and Identification of Crop Disease Based on Depthwise Separable Group Convolution and Feature Fusion

Image recognition technology of crop diseases based on neural network model fusion

A Lightweight Convolutional Neural Network-Based Method for Cotton Mosaic Disease Identification

Rice Disease Identification Method Based on Attention Mechanism and Deep Dense Network