Universal Image Segmentation with Arbitrary Granularity for Efficient Pest Monitoring

Guardat en:
Dades bibliogràfiques
Publicat a:Horticulturae vol. 11, no. 12 (2025), p. 1462-1481
Autor principal: Minh, Dang L
Altres autors: Danish Sufyan, Fayaz Muhammad, Khan, Asma, Arzu, Gul E, Tightiz Lilia, Song Hyoung-Kyu, Moon Hyeonjoon
Publicat:
MDPI AG
Matèries:
Accés en línia:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Etiquetes: Afegir etiqueta
Sense etiquetes, Sigues el primer a etiquetar aquest registre!

MARC

LEADER 00000nab a2200000uu 4500
001 3286299865
003 UK-CbPIL
022 |a 2311-7524 
024 7 |a 10.3390/horticulturae11121462  |2 doi 
035 |a 3286299865 
045 2 |b d20250101  |b d20251231 
100 1 |a Minh, Dang L  |u Institute of Research and Development, Duy Tan University, Da Nang 550000, Vietnam 
245 1 |a Universal Image Segmentation with Arbitrary Granularity for Efficient Pest Monitoring 
260 |b MDPI AG  |c 2025 
513 |a Journal Article 
520 3 |a Accurate and timely pest monitoring is essential for sustainable agriculture and effective crop protection. While recent deep learning-based pest recognition systems have significantly improved accuracy, they are typically trained for fixed label sets and narrowly defined tasks. In this paper, we present RefPestSeg, a universal, language-promptable segmentation model specifically designed for pest monitoring. RefPestSeg can segment targets at any semantic level, such as species, genus, life stage, or damage type, conditioned on flexible natural language instructions. The model adopts a symmetric architecture with self-attention and cross-attention mechanisms to tightly align visual features with language embeddings in a unified feature space. To further enhance performance in challenging field conditions, we integrate an optimized super-resolution module to improve image quality and employ diverse data augmentation strategies to enrich the training distribution. A lightweight postprocessing step refines segmentation masks by suppressing highly overlapping regions and removing noise blobs introduced by cluttered backgrounds. Extensive experiments on a challenging pest dataset show that RefPestSeg achieves an Intersection over Union (IoU) of 69.08 while maintaining robustness in real-world scenarios. By enabling language-guided pest segmentation, RefPestSeg advances toward more intelligent, adaptable monitoring systems that can respond to real-time agricultural demands without costly model retraining. 
653 |a Language 
653 |a Plant protection 
653 |a Datasets 
653 |a Image resolution 
653 |a Optimization techniques 
653 |a Attention 
653 |a Architecture 
653 |a Monitoring systems 
653 |a Image processing 
653 |a Monitoring 
653 |a Machine learning 
653 |a Localization 
653 |a Deep learning 
653 |a Natural language 
653 |a Sustainable agriculture 
653 |a Agriculture 
653 |a Data augmentation 
653 |a Pests 
653 |a Image segmentation 
653 |a Linguistics 
653 |a Image quality 
653 |a Real time 
653 |a Natural language processing 
653 |a Morphology 
653 |a Semantics 
653 |a Environmental 
700 1 |a Danish Sufyan  |u Department of Computer Science and Engineering, Sejong University, Seoul 05006, Republic of Korea; sufyandanish@sju.ac.kr (S.D.); muhammadfayaz@sju.ac.kr (M.F.); asmakhan28@sju.ac.kr (A.K.); arzurabani@sju.ac.kr (G.E.A.); liliatightiz@sejong.ac.kr (L.T.) 
700 1 |a Fayaz Muhammad  |u Department of Computer Science and Engineering, Sejong University, Seoul 05006, Republic of Korea; sufyandanish@sju.ac.kr (S.D.); muhammadfayaz@sju.ac.kr (M.F.); asmakhan28@sju.ac.kr (A.K.); arzurabani@sju.ac.kr (G.E.A.); liliatightiz@sejong.ac.kr (L.T.) 
700 1 |a Khan, Asma  |u Department of Computer Science and Engineering, Sejong University, Seoul 05006, Republic of Korea; sufyandanish@sju.ac.kr (S.D.); muhammadfayaz@sju.ac.kr (M.F.); asmakhan28@sju.ac.kr (A.K.); arzurabani@sju.ac.kr (G.E.A.); liliatightiz@sejong.ac.kr (L.T.) 
700 1 |a Arzu, Gul E  |u Department of Computer Science and Engineering, Sejong University, Seoul 05006, Republic of Korea; sufyandanish@sju.ac.kr (S.D.); muhammadfayaz@sju.ac.kr (M.F.); asmakhan28@sju.ac.kr (A.K.); arzurabani@sju.ac.kr (G.E.A.); liliatightiz@sejong.ac.kr (L.T.) 
700 1 |a Tightiz Lilia  |u Department of Computer Science and Engineering, Sejong University, Seoul 05006, Republic of Korea; sufyandanish@sju.ac.kr (S.D.); muhammadfayaz@sju.ac.kr (M.F.); asmakhan28@sju.ac.kr (A.K.); arzurabani@sju.ac.kr (G.E.A.); liliatightiz@sejong.ac.kr (L.T.) 
700 1 |a Song Hyoung-Kyu  |u Department of Information and Communication Engineering and Convergence Engineering for Intelligent Drone, Sejong University, Seoul 05006, Republic of Korea; songhk@sejong.ac.kr 
700 1 |a Moon Hyeonjoon  |u Department of Computer Science and Engineering, Sejong University, Seoul 05006, Republic of Korea; sufyandanish@sju.ac.kr (S.D.); muhammadfayaz@sju.ac.kr (M.F.); asmakhan28@sju.ac.kr (A.K.); arzurabani@sju.ac.kr (G.E.A.); liliatightiz@sejong.ac.kr (L.T.) 
773 0 |t Horticulturae  |g vol. 11, no. 12 (2025), p. 1462-1481 
786 0 |d ProQuest  |t Agriculture Science Database 
856 4 1 |3 Citation/Abstract  |u https://www.proquest.com/docview/3286299865/abstract/embedded/Q8Z64E4HU3OH5N8U?source=fedsrch 
856 4 0 |3 Full Text + Graphics  |u https://www.proquest.com/docview/3286299865/fulltextwithgraphics/embedded/Q8Z64E4HU3OH5N8U?source=fedsrch 
856 4 0 |3 Full Text - PDF  |u https://www.proquest.com/docview/3286299865/fulltextPDF/embedded/Q8Z64E4HU3OH5N8U?source=fedsrch