相关实验视频

Updated: Sep 11, 2025

Author Spotlight: High-Throughput In Vivo Leaf Inoculation for Accelerating Disease Resistance Screening in Poplar Hybrid Breeding

Author Spotlight: High-Throughput In Vivo Leaf Inoculation for Accelerating Disease Resistance Screening in Poplar Hybrid Breeding

Published on: September 20, 2024

ALD-YOLO:一种轻量级的注意力检测模型,用于检测果叶病

Hong Deng¹, Yiyi Chen¹, Yilu Xu¹

¹School of Software, Jiangxi Agricultural University, Nanchang, China.

Frontiers in plant science

|August 14, 2025

概括

一个新的轻量级模型,ALD-YOLO,增强了果叶病的检测. 它提高了准确性和效率,为实时应用程序和边缘设备提供更快的处理速度.

科学领域:

农业科学农业科学
计算机视觉计算机视觉
机器学习机器学习

背景情况:

果种植在经济上至关重要,但容易受到引起大量收益和财务损失的疾病的影响.
准确和快速检测果叶病对于有效的管理和缓解策略至关重要.
现有的检测方法可能缺乏大型农业应用所需的效率或准确性.

研究的目的:

开发一种轻量级和高效的深度学习模型,以快速准确地检测果叶病.
优化YOLOv8架构,以提高精度和计算效率方面的性能.
提高模型在识别多尺度特征和小疾病目标方面的能力.

主要方法:

提出了ALD-YOLO,这是一个基于YOLOv8架构的轻量级注意力检测模型.
通过使用FasterNet Block优化YOLOv8的C2F模块来提高效率,引入了Faster_C2F模块.
整合了EMA (高效多尺度注意力) 模块,以增强多尺度特征捕获,并专注于较小的疾病目标,创建了Faster_C2F_EMA模块.
将卷积模块替换为ADown,这是一款轻量化下采样模块,以减少参数并保留图像信息.

主要成果:

在AppleLeaf9数据集上,ALD-YOLO显示了mAP (平均平均精度) 的提高,比YOLOv8n增加1.4%,比YOLOv8s增加0.6%.
在GFLOPs (每秒千兆浮点运算) 中实现了显著的减少:与YOLOv8n相比29.63%和与YOLOv8s相比79.93%.
与YOLOv8s相比,CPU推断测试显示,每秒的改进率提高了119.23%.
该模型在检测精度和计算效率之间实现了最佳平衡.

结论:

ALD-YOLO模型为检测果叶病提供了稳定高效的解决方案.
拟议的模型的轻量化性质和改进的性能使其适合在边缘设备上部署.
通过增强疾病检测能力,ALD-YOLO有助于推进精准农业.

关键词:

在C2F中,它是C2F.这是一个YOLO YOLO.果叶的疾病果叶的疾病注意力机制注意力机制对象检测检测对象检测对象检测

更多相关视频

Unravelling the Function of a Bacterial Effector from a Non-cultivable Plant Pathogen Using a Yeast Two-hybrid Screen

Unravelling the Function of a Bacterial Effector from a Non-cultivable Plant Pathogen Using a Yeast Two-hybrid Screen

Published on: January 20, 2017

Specific and Accurate Detection of the Citrus Greening Pathogen Candidatus liberibacter spp. Using Conventional PCR on Citrus Leaf Tissue Samples

Specific and Accurate Detection of the Citrus Greening Pathogen Candidatus liberibacter spp. Using Conventional PCR on Citrus Leaf Tissue Samples

Published on: June 29, 2018

相关概念视频

Light Acquisition

Light Acquisition

In order to produce glucose, plants need to capture sufficient light energy. Many modern plants have evolved leaves specialized for light acquisition. Leaves can be only millimeters in width or tens of meters wide, depending on the environment. Due to competition for sunlight, evolution has driven the evolution of increasingly larger leaves and taller plants, to avoid shading by their neighbors with contaminant elaboration of root architecture and mechanisms to transport water and nutrients.

Adaptations that Reduce Water Loss

Adaptations that Reduce Water Loss

Though evaporation from plant leaves drives transpiration, it also results in loss of water. Because water is critical for photosynthetic reactions and other cellular processes, evolutionary pressures on plants in different environments have driven the acquisition of adaptations that reduce water loss.