摘要
复杂背景下,不同尺度建筑物的特征差异较大,现有算法对多尺度建筑物分割存在分割不均以及误判等问题。为了解决上述问题,本文设计了一种适应多尺度变化的新型网络结构。首先,针对遥感图像场景提分割精度低的问题,引入坐标注意力机制,嵌入到基础网络中增强上下文信息捕获能力,消除噪声的同时增强网络对于空间特征的提取能力。引入了新型递归残差卷积模块,加深网络层次的同时减少信息丢失,提高特征提取效率。最后,在跳跃连接中引入了空洞空间卷积池化金字塔增大网络感受野,增强有效特征,抑制无用特征。设计系统验证模型的实用性。实验结果表明,本文方法在精确率、召回率、F1-score和IoU指标中比U-Net网络分别提高了3.05%、1.56%、1.3%、3.08%。
Under complex background,the characteristics of buildings at different scales are quite different,and the existing algorithms have problems such as uneven segmentation and misjudgment of multi-scale building segmentation.To solve the above problems,we design a new network structure adapted to scale changes.Firstly,aiming at the problem of low segmentation accuracy in remote sensing image scenes,we introduce and embed a coordinate attention mechanism in the basic network to enhance the context information capture ability,eliminate noise and enhance the network′s ability to extract spatial features.we introduce A new recursive residual convolution module ed to deepen the network layer,reduce information loss,and improve the efficiency of feature extraction.Finally,we introduce a hollow space convolutional pooled pyramid in the hop connection to increase the network receptive field,enhance the effective features,and suppress the useless features.Design the system to verify the usefulness of the model.Experimental results show that the proposed method improves the accuracy,recall,F1-score,and IoU indicators by 3.05%,1.56%,1.3%,and 3.08%compared with the U-Net network,respectively.
作者
闫祎巧
王宏生
赵怀慈
刘鹏飞
Yan Yiqiao;Wang Hongsheng;Zhao Huaici;Liu Pengfei(School of Information Science and Engineering,Shenyang University of Technology,Shenyang 110870,China;Key Laboratory of Optical-Electronics Information Processing,Shenyang Institute of Automation Chinese Academy of Sciences,Shenyang 110016,China)
出处
《电子测量技术》
北大核心
2023年第16期172-178,共7页
Electronic Measurement Technology
关键词
语义分割
建筑物提取
遥感图像
注意力机制
semantic segmentation
building extraction
remote sensing image
attention mechanism
作者简介
闫祎巧,硕士研究生,主要研究方向为图像处理等;王宏生,硕士,副教授,主要研究方向为智能搜索引擎,人工智能;赵怀慈,博士,研究员,博导,主要研究方向为图像处理、复杂系统建模与仿真技术,指挥、控制、通信与信息处理技术。E-mail:hczhao@sia.cn
