Tomato disease object detection method combining prior knowledge attention mechanism and multiscale features

Liu, Jun; Wang, Xuewei

doi:10.3389/fpls.2023.1255119

Cited by 6 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Object detection algorithms can be classified into three primary categories according to their underlying principles: those that depend on prior knowledge, machine learning, and deep learning. Perception algorithms that rely on prior knowledge [ 4 ] generally consist of two distinct stages: hypothesis generation (HG) and regions of interest (ROI) identification. The HG step generates regions within an image likely to contain the target object.…”

Section: Related Workmentioning

confidence: 99%

HRYNet: A Highly Robust YOLO Network for Complex Road Traffic Object Detection

Tang,

Yun,

Chen

et al. 2024

Sensors

View full text Add to dashboard Cite

Object detection is a crucial component of the perception system in autonomous driving. However, the road scene presents a highly intricate environment where the visibility and characteristics of traffic targets are susceptible to attenuation and loss due to various complex road scenarios such as lighting conditions, weather conditions, time of day, background elements, and traffic density. Nevertheless, the current object detection network must exhibit more learning capabilities when detecting such targets. This also exacerbates the loss of features during the feature extraction and fusion process, significantly compromising the network’s detection performance on traffic targets. This paper presents a novel methodology by which to overcome the concerns above, namely HRYNet. Firstly, a dual fusion gradual pyramid structure (DFGPN) is introduced, which employs a two-stage gradient fusion strategy to enhance the generation of more comprehensive multi-scale high-level semantic information, strengthen the interconnection between non-adjacent feature layers, and reduce the information gap that exists between them. HRYNet introduces an anti-interference feature extraction module, the residual multi-head self-attention mechanism (RMA). RMA enhances the target information by implementing a characteristic channel weighting policy, thereby reducing background interference and improving the attention capability of the network. Finally, the detection performance of HRYNet was evaluated by utilizing three datasets: the horizontally collected dataset BDD1000K, the UAV high-altitude dataset Visdrone, and a custom dataset. Experimental results demonstrate that HRYNet achieves a higher mAP_0.5 compared with YOLOv8s on the three datasets, with increases of 10.8%, 16.7%, and 5.5%, respectively. To optimize HRYNet for mobile devices, this study presents Lightweight HRYNet (LHRYNet), which effectively reduces the number of model parameters by 2 million. The results demonstrate that LHRYNet outperforms YOLOv8s in terms of mAP_0.5, with improvements of 6.7%, 10.9%, and 2.5% observed on the three datasets, respectively.

show abstract

Section: Related Workmentioning

confidence: 99%

HRYNet: A Highly Robust YOLO Network for Complex Road Traffic Object Detection

Tang,

Yun,

Chen

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…First, accurate and rapid classification of diseases is an important basis for early disease monitoring, diagnostics, and prevention (Abdullah et al, 2023;Liu and Wang, 2023). In recent years, deep convolutional neural networks have achieved remarkable success in disease image classification (Ning et al, 2023;Zhang et al, 2023).…”

Section: Limitations and Prospectsmentioning

confidence: 99%

IoT-based system of prevention and control for crop diseases and insect pests

Wang,

Qiao,

Wang

et al. 2024

Front. Plant Sci.

View full text Add to dashboard Cite

Environmentally friendly technologies for the prevention and control of crop diseases and insect pests are important to reduce the use of chemical pesticides, improve the quality of agricultural products, protect the environment, and promote sustainable development of crop production. On the basis of Internet of Things (IoT) technology, we developed a prevention and control system for crop diseases and insect pests with two main components: a plant protection device (the hardware) and an information management system (the software). To be suitable for both facility- and field-based production scenarios, we incorporated two types of plant protection devices, utilizing ozone sterilization and light-trap technologies. The devices were equipped with various sensors to realize real-time collection and monitoring of data on the crop production environment. The information management system has an IoT-based architecture and includes a mobile device app to enable remote control of the plant protection devices for intelligent management of plant protection data. The system can achieve efficient management of large-scale equipment applications and multi-device collaborative work to prevent and control pests and diseases. The developed system has operated successfully for several years in China and has been applied to cucumber, tomato, rice, and other crops. We demonstrate the effectiveness and practicality of the system in a greenhouse facility and in the field.

show abstract

“…(2022c) proposed a lightweight detection method that combines improved YOLOv5 and ShuffleNet to detect peach tree leaf diseases in natural environments, albeit with a slight decrease in accuracy. Liu and Wang (2023) introduced a mixed attention mechanism into the feature prediction structure of YOLOv5 to improve the detection of tomato brown spot disease in complex scenes. Zhang et al.…”

Section: Introductionmentioning

confidence: 99%

Tomato leaf disease detection based on attention mechanism and multi-scale feature fusion

Wang,

Zhang,

Tian

2024

Front. Plant Sci.

View full text Add to dashboard Cite

When detecting tomato leaf diseases in natural environments, factors such as changes in lighting, occlusion, and the small size of leaf lesions pose challenges to detection accuracy. Therefore, this study proposes a tomato leaf disease detection method based on attention mechanisms and multi-scale feature fusion. Firstly, the Convolutional Block Attention Module (CBAM) is introduced into the backbone feature extraction network to enhance the ability to extract lesion features and suppress the effects of environmental interference. Secondly, shallow feature maps are introduced into the re-parameterized generalized feature pyramid network (RepGFPN), constructing a new multi-scale re-parameterized generalized feature fusion module (BiRepGFPN) to enhance feature fusion expression and improve the localization ability for small lesion features. Finally, the BiRepGFPN replaces the Path Aggregation Feature Pyramid Network (PAFPN) in the YOLOv6 model to achieve effective fusion of deep semantic and shallow spatial information. Experimental results indicate that, when evaluated on the publicly available PlantDoc dataset, the model’s mean average precision (mAP) showed improvements of 7.7%, 11.8%, 3.4%, 5.7%, 4.3%, and 2.6% compared to YOLOX, YOLOv5, YOLOv6, YOLOv6-s, YOLOv7, and YOLOv8, respectively. When evaluated on the tomato leaf disease dataset, the model demonstrated a precision of 92.9%, a recall rate of 95.2%, an F1 score of 94.0%, and a mean average precision (mAP) of 93.8%, showing improvements of 2.3%, 4.0%, 3.1%, and 2.7% respectively compared to the baseline model. These results indicate that the proposed detection method possesses significant detection performance and generalization capabilities.

show abstract

Tomato disease object detection method combining prior knowledge attention mechanism and multiscale features

Cited by 6 publications

References 31 publications

HRYNet: A Highly Robust YOLO Network for Complex Road Traffic Object Detection

HRYNet: A Highly Robust YOLO Network for Complex Road Traffic Object Detection

IoT-based system of prevention and control for crop diseases and insect pests

Tomato leaf disease detection based on attention mechanism and multi-scale feature fusion

Contact Info

Product

Resources

About