基于改进Unet模型的无人机影像两阶段草地退化指示物种分类  

Two-stage Grassland Degradation Indicator Species Classification based on Improved Unet Model for UAV Images

作　　者：陈志程 万华伟 万凤鸣 高吉喜 孙林[2] 杨斌 Zhicheng CHEN;Huawei WAN;Fengming WAN;Jixi GAO;Lin SUN;Bin YANG(Center for Satellite Application on Ecology and Environment,Ministry of Ecology and Environment,Beijing 100094,China;School of Geomatics and Spatial Information,Shandong University of Science and Technology,Qingdao 266000,China;Chinese Research Academy of Environmental Sciences,Beijing 100012,China;YUSENSE Information Technology and Equipment(Qingdao)Limited Liablity Company,Qingdao 266000,China)

机构地区：[1]生态环境部卫星环境应用中心,北京100094 [2]山东科技大学测绘与空间信息学院,山东青岛266000 [3]中国环境科学研究院,北京100012 [4]山东长光禹辰信息技术与装备(青岛)有限公司,山东青岛266000

出　　处：《遥感技术与应用》2024年第5期1075-1084,共10页Remote Sensing Technology and Application

基　　金：国家重点研发计划项目“草地生物多样性无伤害遥感监测技术与应用示范”(2021YFB3901102)。

摘　　要：针对退化指示物种植株体积小、草种间形态特征相似易造成混合像元等问题,根据所获取低空无人机数据,提出一种基于目标检测和语义分割的两阶段分类方法,其次对于分割模型进行轻量化改进。采用结构重参数化RepVGG网络替代Unet主干网络,在编码阶段导入高效通道注意力机制ECA,在下采样环节提升模型的特征提取能力,实现轻量化特征提取,块结构使用ESE模块,避免通道信息的损失。改进后的分割模型对于锡林浩特典型草原的冷蒿和银灰旋花两类草地退化指示物种有很好的分类效果,MIoU可以达到0.91,相比原始Unet模型提升0.11左右。实验结果表明:无人机数据以及两阶段分类方法可以很好地进行草地退化指示物种分类,提出的轻量化改进模型效果良好。Aiming at the problems of small plant size of degraded indicator species and mixed pixels caused by similar morphological characteristics between grass species,a two-stage classification method based on object detection and semantic segmentation is proposed according to the obtained low-altitude UAV data.Secondly,the segmentation model is lightweight improved.The RepVGG network with structural reparameterization is used to replace the Unet backbone network.The efficient channel attention mechanism ECA is introduced in the coding stage,and the feature extraction ability of the model is improved in the down-sampling link to achieve lightweight feature extraction.The block structure uses the ESE module to avoid the loss of channel information.The improved segmentation model has a good classification effect on the two types of grassland degradation indicator species of Artemisia frigida and Convolvulus ammannii in the typical grassland of Xilinhot.The MIoU can reach 0.91,which is about 0.11 higher than the original Unet model.The experimental results show that the UAV data and the two-stage classification method can classify the grassland degradation indicator species well,and the proposed lightweight improved model has a good effect.

关 键 词：草地退化指示物种 目标检测 语义分割 两阶段 轻量化 分类 

分 类 号：S812[农业科学—草业科学] TP79[农业科学—畜牧学]