Potato late blight leaf detection in complex environments DOI Creative Commons
Jingtao Li,

Jiawei Wu,

Rui Liu

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Дек. 28, 2024

Abstract Potato late blight is a common disease affecting crops worldwide. To help detect this in complex environments, an improved YOLOv5 algorithm proposed. First, ShuffleNetV2 used as the backbone network to reduce number of parameters and computational load, making model more lightweight. Second, coordinate attention mechanism added missed detection for leaves that are overlapping, damaged, or hidden, thereby increasing accuracy under challenging conditions. Lastly, bidirectional feature pyramid employed fuse information different scales. The study results show significant improvement model’s performance. was reduced from 7.02 3.87 M, floating point operations dropped 15.94 8.4 G. These reductions make lighter efficient. speed increased by 16 %, enabling faster potato leaves. Additionally, average precision 3.22 indicating better accuracy. Overall, provides robust solution detecting environments. study’s findings can be useful applications further research controlling similar

Язык: Английский

A Multi-Scale Feature Focus and Dynamic Sampling-Based Model for Hemerocallis fulva Leaf Disease Detection DOI Creative Commons

Tiebiao Wang,

H. Xia, Jiao Xie

и другие.

Agriculture, Год журнала: 2025, Номер 15(3), С. 262 - 262

Опубликована: Янв. 25, 2025

Hemerocallis fulva, essential to urban ecosystems and landscape design, faces challenges in disease detection due limited data reduced accuracy complex backgrounds. To address these issues, the fulva leaf dataset (HFLD-Dataset) is introduced, alongside Multi-Scale Enhanced Network (HF-MSENet), an efficient model designed improve multi-scale reduce misdetections. The Channel–Spatial Module (CSMSM) enhances localization capture of critical features, overcoming limitations feature extraction caused by inadequate attention characteristics. C3_EMSCP module improves fusion combining convolutional kernels group convolution, increasing adaptability interaction across scales. interpolation errors boundary blurring upsampling, DySample adapts sampling positions using a dynamic offset learning mechanism. This, combined with pixel reordering grid techniques, reduces preserves edge details. Experimental results show that HF-MSENet achieves mAP@50 mAP%50–95 scores 94.9% 80.3%, respectively, outperforming baseline 1.8% 6.5%. Compared other models, demonstrates significant advantages efficiency robustness, offering reliable support for precise fulva.

Язык: Английский

Процитировано

0
YOLO-BSMamba: A YOLOv8s-Based Model for Tomato Leaf Disease Detection in Complex Backgrounds DOI Creative Commons
Zhihong Liu, Xiangyun Guo, Tian Zhao

и другие.

Agronomy, Год журнала: 2025, Номер 15(4), С. 870 - 870

Опубликована: Март 30, 2025

The precise identification of diseases in tomato leaves is great importance for target pesticide application a complex background scenario. Existing models often have difficulty capturing long-range dependencies and fine-grained features images, leading to poor recognition where there are backgrounds. To tackle this challenge, study proposed using YOLO-BSMamba detection mode. We that Hybrid Convolutional Mamba module (HCMamba) integrated within the neck network, with aim improving feature representation by leveraging capture global contextual capabilities State Space Model (SSM) discerning localized spatial convolution. Furthermore, we introduced Similarity-Based Attention Mechanism into C2f improve model’s extraction focusing on disease-indicative leaf areas reducing noise. weighted bidirectional pyramid network (BiFPN) was utilized replace feature-fusion component thereby enhancing performance lesions exhibiting heterogeneous symptomatic gradations enabling model effectively integrate from different scales. Research results showed achieved an F1 score, [email protected], [email protected]:0.95 81.9%, 86.7%, 72.0%, respectively, which represents improvement 3.0%, 4.8%, 4.3%, compared YOLOv8s. Compared other YOLO series models, it achieves best [email protected] score. This provides robust reliable method disease recognition, expected efficiency, further enhance crop monitoring management precision agriculture.

Язык: Английский

Процитировано

0
CEFW-YOLO: A High-Precision Model for Plant Leaf Disease Detection in Natural Environments DOI Creative Commons
J. Tao, Xiaoli Li, Yong He

и другие.

Agriculture, Год журнала: 2025, Номер 15(8), С. 833 - 833

Опубликована: Апрель 12, 2025

The accurate and rapid detection of apple leaf diseases is a critical component precision management in orchards. existing deep-learning-based algorithms for typically demand high computational resources, which limits their practical applicability orchard environments. Furthermore, the natural settings faces significant challenges due to diversity disease types, varied morphology affected areas, influence factors such as lighting variations, occlusions, differences severity. To address above challenges, we constructed an (ALD) dataset, was collected from real-world scenarios, applied data augmentation techniques, resulting total 9808 images. Based on ALD proposed lightweight YOLO11n-based network, named CEFW-YOLO, designed tackle current issues identification. First, novel channel-wise squeeze convolution (CWSConv), employs channel compression standard reduce resource consumption, enhance small objects, improve model’s adaptability morphological complex backgrounds. Second, developed enhanced cross-channel attention (ECCAttention) module integrated it into C2PSA_ECCAttention module. By extracting global information, combining horizontal vertical convolutions, strengthening interactions, this enables model more accurately capture features leaves, thereby enhancing accuracy robustness. Additionally, introduced new fine-grained multi-level linear (FMLAttention) module, utilizes asymmetric convolutions mechanisms ability local details detection. Finally, incorporated Wise-IoU (WIoU) loss function, enhances differentiate overlapping targets across multiple scales. A comprehensive evaluation CEFW-YOLO conducted, comparing its performance against state-of-the-art (SOTA) models. achieved 20.6% reduction complexity. Compared original YOLO11n, improved by 3.7%, with [email protected] [email protected]:0.95 increasing 7.6% 5.2%, respectively. Notably, outperformed advanced SOTA detection, underscoring application potential scenarios.

Язык: Английский

Процитировано

0
Potato late blight leaf detection in complex environments DOI Creative Commons
Jingtao Li,

Jiawei Wu,

Rui Liu

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Дек. 28, 2024

Abstract Potato late blight is a common disease affecting crops worldwide. To help detect this in complex environments, an improved YOLOv5 algorithm proposed. First, ShuffleNetV2 used as the backbone network to reduce number of parameters and computational load, making model more lightweight. Second, coordinate attention mechanism added missed detection for leaves that are overlapping, damaged, or hidden, thereby increasing accuracy under challenging conditions. Lastly, bidirectional feature pyramid employed fuse information different scales. The study results show significant improvement model’s performance. was reduced from 7.02 3.87 M, floating point operations dropped 15.94 8.4 G. These reductions make lighter efficient. speed increased by 16 %, enabling faster potato leaves. Additionally, average precision 3.22 indicating better accuracy. Overall, provides robust solution detecting environments. study’s findings can be useful applications further research controlling similar

Язык: Английский

Процитировано

0