A multimodal framework for pepper diseases and pests detection

Abstract Pepper diseases and pests typically exhibit small target proportions, diverse shapes and sizes, complex imaging backgrounds, and similarities with the background. Existing detection methods perform poorly in identifying targets of different sizes and shapes within the same scene, and they l...

Full description

Saved in:
Bibliographic Details
Main Authors: Jun Liu, Xuewei Wang
Format: Article
Language:English
Published: Nature Portfolio 2024-11-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-024-80675-w
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846158613313224704
author Jun Liu
Xuewei Wang
author_facet Jun Liu
Xuewei Wang
author_sort Jun Liu
collection DOAJ
description Abstract Pepper diseases and pests typically exhibit small target proportions, diverse shapes and sizes, complex imaging backgrounds, and similarities with the background. Existing detection methods perform poorly in identifying targets of different sizes and shapes within the same scene, and they lack adequate noise suppression capabilities. To address the practical needs of detecting pepper diseases and pests in complex scenarios, we have constructed the first multimodal pepper diseases and pests object detection dataset (PDD). This dataset includes a wide variety of diseases and pests images, along with detailed natural language descriptions of their attributes. Locating the described targets in complex scenes with similar disease symptoms and leaf occlusion presents a significant challenge. To tackle this issue, we propose the PepperNet model for object detection in pepper diseases and pests images using natural language descriptions. This model decomposes complex multimodal features of language and images into explicit attribute features and employs fine-grained multimodal attribute contrast learning strategies. This approach effectively distinguishes subtle local differences between similar objects, achieving fine-grained mapping from language to vision in complex scenarios. Our detection results show a mAP@0.5 of 91.93% and a detection speed of 121.8 frames per second. Visualizations indicate that the model maintains high robustness under varying noise levels and occlusion conditions, demonstrating superior performance and stability across diverse complex scenarios.
format Article
id doaj-art-a49c63fabdae45159e694c8e923db62d
institution Kabale University
issn 2045-2322
language English
publishDate 2024-11-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-a49c63fabdae45159e694c8e923db62d2024-11-24T12:22:55ZengNature PortfolioScientific Reports2045-23222024-11-0114112010.1038/s41598-024-80675-wA multimodal framework for pepper diseases and pests detectionJun Liu0Xuewei Wang1Shandong Provincial University Laboratory for Protected Horticulture, Weifang University of Science and TechnologyShandong Provincial University Laboratory for Protected Horticulture, Weifang University of Science and TechnologyAbstract Pepper diseases and pests typically exhibit small target proportions, diverse shapes and sizes, complex imaging backgrounds, and similarities with the background. Existing detection methods perform poorly in identifying targets of different sizes and shapes within the same scene, and they lack adequate noise suppression capabilities. To address the practical needs of detecting pepper diseases and pests in complex scenarios, we have constructed the first multimodal pepper diseases and pests object detection dataset (PDD). This dataset includes a wide variety of diseases and pests images, along with detailed natural language descriptions of their attributes. Locating the described targets in complex scenes with similar disease symptoms and leaf occlusion presents a significant challenge. To tackle this issue, we propose the PepperNet model for object detection in pepper diseases and pests images using natural language descriptions. This model decomposes complex multimodal features of language and images into explicit attribute features and employs fine-grained multimodal attribute contrast learning strategies. This approach effectively distinguishes subtle local differences between similar objects, achieving fine-grained mapping from language to vision in complex scenarios. Our detection results show a mAP@0.5 of 91.93% and a detection speed of 121.8 frames per second. Visualizations indicate that the model maintains high robustness under varying noise levels and occlusion conditions, demonstrating superior performance and stability across diverse complex scenarios.https://doi.org/10.1038/s41598-024-80675-wObject detectionPepper diseases and pests imageNatural LanguageMultimodalVisual features
spellingShingle Jun Liu
Xuewei Wang
A multimodal framework for pepper diseases and pests detection
Scientific Reports
Object detection
Pepper diseases and pests image
Natural Language
Multimodal
Visual features
title A multimodal framework for pepper diseases and pests detection
title_full A multimodal framework for pepper diseases and pests detection
title_fullStr A multimodal framework for pepper diseases and pests detection
title_full_unstemmed A multimodal framework for pepper diseases and pests detection
title_short A multimodal framework for pepper diseases and pests detection
title_sort multimodal framework for pepper diseases and pests detection
topic Object detection
Pepper diseases and pests image
Natural Language
Multimodal
Visual features
url https://doi.org/10.1038/s41598-024-80675-w
work_keys_str_mv AT junliu amultimodalframeworkforpepperdiseasesandpestsdetection
AT xueweiwang amultimodalframeworkforpepperdiseasesandpestsdetection
AT junliu multimodalframeworkforpepperdiseasesandpestsdetection
AT xueweiwang multimodalframeworkforpepperdiseasesandpestsdetection