Swin Attention Augmented Residual Network: a fine-grained pest image recognition method.

Front Plant Sci

School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, China.

Published: June 2025


Category Ranking

98%

Total Visits

921

Avg Visit Duration

2 minutes

Citations

20

Article Abstract

Pest infestation is a major cause of crop losses and a significant factor contributing to agricultural economic damage. Accurate identification of pests is therefore critical to ensuring crop safety. However, existing pest recognition methods often struggle to distinguish fine-grained visual differences between pest species and are susceptible to background interference from crops and environments. To address these challenges, we propose an improved pest identification method based on the Swin Transformer architecture, named Swin-AARNet (Attention Augmented Residual Network). Our method achieves efficient and accurate pest recognition. On the one hand, Swin-AARNet enhances local key features and establishes a feature complementation mechanism, thereby improving the extraction capability of local features. On the other hand, it integrates multi-scale information to effectively alleviate the problem of fine-grained feature ambiguity or loss. Furthermore, Swin-AARNet attained a classification accuracy of 78.77% on IP102, the largest publicly available pest dataset to date. To further validate its effectiveness and generalization ability, we conducted additional training and evaluation on the citrus benchmark dataset CPB and Li, achieving impressive accuracies of 82.17% and 99.48%, respectively. SwinAARNet demonstrates strong capability in distinguishing pests with highly similar appearances while remaining robust against complex and variable backgrounds. This makes it a promising tool for enhancing agricultural safety management, including crop environment monitoring and early invasion warning. Compared with other state-of-the-art models, our proposed method exhibits superior performance in pest image classification tasks, highlighting its potential for real-world agricultural applications.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC12222059PMC
http://dx.doi.org/10.3389/fpls.2025.1619551DOI Listing

Publication Analysis

Top Keywords

attention augmented
8
augmented residual
8
residual network
8
pest
8
pest image
8
pest recognition
8
swin attention
4
network fine-grained
4
fine-grained pest
4
image recognition
4

Similar Publications

ObjectiveThis work examined performance costs for a spatial integration task when two sources of information were presented at increasing eccentricities with an augmented-reality (AR) head-mounted display (HMD).BackgroundSeveral studies have noted that different types of tasks have varying costs associated with the spatial proximity of information that requires mental integration. Additionally, prior work has found a relatively negligible role of head movements associated with performance costs.

View Article and Find Full Text PDF

Crohn's disease pathology is modeled in TNF mice that overproduce tumor necrosis factor (TNF) to drive disease through TNF receptors. An alternative ligand for TNF receptors, soluble LTα, is produced by B cells, but has received scarce attention because LTα also partners with LTβ to generate membrane-tethered LTαβ that promotes tertiary lymphoid tissue-another feature of Crohn's disease. We hypothesized that B cell-derived LTαβ would critically affect ileitis in TNF mice.

View Article and Find Full Text PDF

Non-intrusive neuroimaging technology offers fast and robust diagnostic tools for neuro-disorder disease diagnosis, such as Attention-Deficit/Hyperactivity Disorder (ADHD). Resting-state functional magnetic imaging (rs-fMRI) has been demonstrated to have great potential for such applications due to its unique capability and convenience in providing spatial-temporal brain imaging. One critical challenge of using rs-fMRI data is the high dimensionality for both spatial and temporal domains.

View Article and Find Full Text PDF

Objectives: We propose a YOLOv11-TDSP model for improving the accuracy of dental abnormality detection on panoramic oral X-ray images.

Methods: The SHSA single-head attention mechanism was integrated with C2PSA in the backbone layer to construct a new C2PSA_SHSA attention mechanism. The computational redundancy was reduced by applying single-head attention to some input channels to enhance the efficiency and detection accuracy of the model.

View Article and Find Full Text PDF

Objective: This study aims to develop a robust, multi-task deep learning framework that integrates vessel segmentation and radiomic analysis for the automated classification of four retinal conditions- diabetic retinopathy (DR), hypertensive retinopathy (HR), papilledema, and normal fundus-using fundus images.

Materials: AND.

Methods: A total of 2,165 patients from eight medical centers were enrolled.

View Article and Find Full Text PDF