GBR-DETR：一种用于边缘设备部署的实时番茄叶片病害检测模型
GBR-DETR: A Real-Time Tomato Leaf Disease Detection Model for Edge Device Deployment.
摘要(中文译文)
番茄叶片病害对作物产量和粮食安全构成重大威胁。然而，在真实栽培环境中，光照波动、叶片遮挡程度不一以及病斑形态模糊等因素往往会影响检测精度。本文提出了一种梯度感知双向保留检测Transformer（GBR-DETR），该模型专为高精度、实时病害检测而设计。该模型由两个网络结构和一个保留特征聚合模块组成：（1）设计了多尺度梯度感知迁移网络（MGAT-Net），通过Sobel算子编码梯度信息，从而增强对小目标和模糊病斑的定位稳定性；（2）提出了双向上下文金字塔网络（BCPN），通过自顶向下和自底向上的路径实现多层级特征之间的双向交互，从而生成多尺度病斑特征并弥合跨尺度语义鸿沟；（3）采用保留特征聚合模块（RFAM）抑制背景噪声并建立全局特征关联，从而增强病斑识别的整体表征能力。在多场景番茄叶片病害（M-TLD）数据集上的实验表明，GBR-DETR相较于基线RT-DETR在mAP、mAP和mAP上分别提升了3.12、4.88和3.41个百分点，同时优于代表性的基于DETR和基于CNN的检测器。该模型在PlantDoc跨域基准上展现出稳健的泛化能力，mAP较基线提升2.11%。在NVIDIA Jetson Orin Nano上以TensorRT FP16部署时，实现了54 ms的延迟，可在边缘设备上实现实时病害监测。该方案为智慧农业中的实时病害监测提供了有效的技术支持。
Abstract
Tomato leaf diseases pose significant threats to crop yield and food security. However, in real-world cultivation environments, factors such as fluctuating illumination, varying leaf occlusion, and ambiguous lesion morphology often compromise detection accuracy. This paper presents the Gradient-aware Bidirectional Retentive Detection Transformer (GBR-DETR), a model designed for high-precision, real-time disease detection. This model is composed of two network structures and a retentive feature aggregation module: (1) a Multi-scale Gradient-Aware Transfer Network (MGAT-Net) is designed to encode gradient information through the Sobel operator, thereby enhancing the localization stability for small and blurry lesions; (2) a Bidirectional Context Pyramid Network (BCPN) is proposed to enable bidirectional interactions among multi-level features through a top-down and a bottom-up pathway, thereby generating multi-scale lesion features and bridging cross-scale semantic gaps; and (3) a Retentive Feature Aggregation Module (RFAM) is used to suppress background noise and establish global feature correlations, thereby enhancing the overall representation capability for lesion recognition. Experiments on the Multi-scenario Tomato Leaf Disease (M-TLD) dataset show that GBR-DETR yields gains of 3.12, 4.88, and 3.41 percentage points in mAP, mAP, and mAP, respectively, over the baseline RT-DETR, while also outperforming representative DETR-based and CNN-based detectors. The model demonstrates robust generalization on the PlantDoc cross-domain benchmark, achieving a 2.11% improvement in mAP over the baseline. Deployed on the NVIDIA Jetson Orin Nano with TensorRT FP16, it achieves 54 ms latency, enabling real-time disease monitoring on edge devices. This solution provides effective technical support for real-time disease monitoring in smart agriculture.
如何引用
AMA
Jiaxiong Zhuo, Guikun Dong, Qingfeng Huang, Lei Zhou, Feixiong Zhao, Ping Yuan, et al. GBR-DETR: A Real-Time Tomato Leaf Disease Detection Model for Edge Device Deployment.. Sensors (Basel, Switzerland). 2026; doi:10.3390/s26102950.
APA
Jiaxiong Zhuo, Guikun Dong, Qingfeng Huang, Lei Zhou, Feixiong Zhao, Ping Yuan, et al (2026). GBR-DETR: A Real-Time Tomato Leaf Disease Detection Model for Edge Device Deployment.. Sensors (Basel, Switzerland). https://doi.org/10.3390/s26102950
GB/T 7714
Jiaxiong Zhuo, Guikun Dong, Qingfeng Huang, Lei Zhou, Feixiong Zhao, Ping Yuan, et al. GBR-DETR: A Real-Time Tomato Leaf Disease Detection Model for Edge Device Deployment.[J]. Sensors (Basel, Switzerland), 2026 doi:10.3390/s26102950.
在 PubMed增强版 阅读全文
翻译全文 · AI 文献问答 · PDF 原文下载 · 引用导出