融合注意力机制和权重平衡算法的遥感图像分类  

Remote Sensing Image Classification Based on Fusion of Attention Mechanism and Weight Balance Algorithm

作　　者：王民水 王明常[1] 王婧瑜[2] 刘子维 Wang Minshui;Wang Mingchang;Wang Jingyu;Liu Ziwei(College of GeoExploration Science and Technology,Jilin University,Changchun 130026,China;Jilin Academy of Agricultural Sciences,Changchun 130033,China)

机构地区：[1]吉林大学地球探测科学与技术学院,长春130026 [2]吉林省农业科学院,长春130033

出　　处：《吉林大学学报(地球科学版)》2025年第2期697-704,共8页Journal of Jilin University:Earth Science Edition

基　　金：国家自然科学基金项目(42171407)。

摘　　要：针对城市遥感图像各种地物分布不均衡、分类精度较低的问题,提出融合并行注意力与权重平衡算法的遥感图像分类方法。该方法在DeepLabV3+和ResNet50创建的语义分割网络基础上,采用并行组合方式,融入通道注意力和空间注意力算法,提高网络的特征提取能力;针对遥感图像地物类别占比不均衡问题,引入地物类别权重平衡算法,提高小类别地物的分类精度。为了验证网络模型的分类效果,利用Vaihingen数据集和Postdam数据集进行实验。实验结果表明:融合注意力机制和权重平衡算法的分类网络在Vaihingen数据集中测试数据的像素精度、平均交并比、平均F_(1)值分别为96.66%、90.35%、96.66%,在Postdam数据集中测试数据的像素精度、平均交并比、平均F_(1)值分别为95.74%、81.47%、91.82%;从分类细节看,增加注意力机制和权重平衡算法对占比较少的汽车识别精度有显著提高,在Vaihingen数据集中汽车的像素精度提高了26.44%,在Postdam数据集中汽车的像素精度提高了21.84%,取得了较好的分类效果。Addressing the challenge posed by the uneven distribution of various features and the low classification accuracy of urban remote sensing images,we propose a novel method for remote sensing image classification that integrates parallel attention and weight balance algorithm.Leveraging the semantic segmentation network architecture of DeepLabV3+and ResNet50,our method combines channel attention and spatial attention algorithms in parallel to improve the network's feature extraction capability.Additionally,to address the issue of imbalanced feature category proportions in remote sensing images,we propose a feature category weight balance algorithm to improve the classification accuracy of minority feature categories.To validate the effectiveness of our network model for classification,we conduct experiments using Vaihingen and Postdam datasets.The experimental results demonstrate promising performance metrics:The remote sensing image classification algorithm that integrates attention mechanism and weight balance is validated in the Vaihingen dataset with pixel accuracy,mean intersection over union,and mean F_(1) values of 96.66%,90.35%,and 96.66%,respectively.In the Postdam dataset,the pixel accuracy,mean intersection over union,and mean F_(1) values of the validated data are 95.74%,81.47%,and 91.82%,respectively.From the classification details,incorporating an attention mechanism and a weight balance algorithm significantly enhances the recognition accuracy of cars,which account for a relatively small proportion.Specifically,the pixel accuracy of cars in Vaihingen dataset has improved by 26.44%,and in Postdam dataset,it has increased by 21.84%,leading to commendable classification results.

关 键 词：注意力机制 权重平衡算法 DeepLabV3+网络 遥感图像 地物分类 

分 类 号：P79[天文地球—海洋科学]