山东农业科学2025，Vol.57Issue(4)：136-145,10.DOI:10.14083/j.issn.1001-4942.2025.04.017
基于卫星遥感影像的茶园精准识别研究
Research on Accurate Identification of Tea Gardens Based on Satellite Remote Sensing Images
摘要
Abstract
Due to the complex and varied textural features and spatial distribution characteristics of tea garden in satellite remote sensing images,it is a challenge to effectively recognize tea garden using traditional methods.Therefore,this study was aimed to achieve precise identification of tea gardens through satellite re-mote sensing images combined with high-precision semantic segmentation technology.Firstly,according to the satellite images of Rizhao City,the Labelme tool was utilized to accurately annotate the textural information of tea gardens and construct a dataset of tea garden remote sensing images.Based on this dataset,seven semantic segmentation models such as Vgg-unet,Resnet50-unet and Segformer_b2 were trained using data augmentation techniques for precise identification of tea gardens.The results indicated that among the seven models,Seg-former_b2 showed the best property with accuracy of 91.29％,recall rate of 91.09％,mean intersection over union(mIoU)of 84.45％and F1-score of 91.20％for tea garden remote sensing image segmentation.To fur-ther enhance the performance of the Segformer_b2 model in identifying tea gardens,the multi-head context at-tention(MCA)mechanism was introduced,and strided convolution was integrated to enhance the model's a-bility to capture key textural features of tea gardens and its perception and processing capabilities for local de-tails in the images.The improved Segformer-b2 model exhibited enhanced performance in tea garden remote sensing image segmentation with accuracy increased to 91.31％,recall rate increased to 92.97％,mIoU in-creased to 85.87％,and F1-score increased to 92.12％,which could achieve the high precision and efficiency identification of tea gardens remote sensing images.This study could not only provide strong support for future precise identification and management of tea gardens,but also showcase the potential of satellite remote sens-ing technology in the agricultural field.关键词
卫星遥感影像/茶园识别/语义分割模型Key words
Satellite remote sensing images/Tea garden recognization/Semantic segmentation model分类
农业科技引用本文复制引用
李冬雪,李峰,赵梦洁,丁兆堂,马青平..基于卫星遥感影像的茶园精准识别研究[J].山东农业科学,2025,57(4):136-145,10.基金项目
国家自然科学基金青年基金项目(32302607) （32302607）
山东省高等学校"青创团队计划"项目(2023KJ208) （2023KJ208）
聊城大学风景园林学科开放课题(319462212) （319462212）
