| 引用本文: | 汪 强,郭来功,程伟涛.基于轻量化 NDFEDet-SOLOv2 的遥感图像建筑物提取方法(J/M/D/N,J:杂志，M：书，D：论文，N：报纸).期刊名称,2024，41（6）：20-29 |
| CHEN X. Adap tive slidingmode contr ol for discrete2ti me multi2inputmulti2 out put systems[ J ]. Aut omatica, 2006, 42(6): 4272-435 |
|
| 摘要: |
| 目的 在地籍测绘和灾害管理等领域中,建筑物轮廓和位置的自动提取是至关重要的一环。 为了解决高分
辨率遥感图像建筑物因环境因素导致分割精度不准确等问题,提出了一种改进的轻量化 SOLOv2 实例分割模
型———NDFEDet-SOLOv2。 方法 该模型选用双向特征金字塔网络( BiFPN) 特征融合方式的轻量级 EfficientDet 网
络,其中将骨干网络部分的 EfficientNet 升级为 EfficientNetv2,EfficientNetv2 中的三层 MBConv 模块 SE 注意力更换
为含有 DropBlock 正则化的轻量级标准化注意力机制( NAM) ,构成 NAD-MBConv 模块。 BiFPN 特征融合部分,向
其尾端各特征层并入双水平路由注意视觉变压器( BiFormer) ,形成双向水平路由注意特征金字塔网络结构( Bi-
FPN-Former) ,从而聚焦微小建筑物轮廓信息,以实现更高层次的特征融合。 结果 NDFEDet-SOLOv2 模型相较于传
统轻量级 SOLOv2 实例分割算法,平均精度 mAP、mAP50 和 mAP75 分别提高了 3. 9%、3. 7%和 2. 5%,检测帧率(FPS)提
高了 2. 7 帧 / s。 结论 轻量化 NDFEDet-SOLOv2 实例分割算模型消除了建筑物边角的图像畸变,在地理环境空间不均
等复杂情况下也能准确提取出遥感图像建筑物的基本轮廓,从而为城市布局更新和建筑变化检测提供理论参考。 |
| 关键词: 高分 辨 率 遥 感 图 像 实 例 分 割 EfficientDet 标 准 化 注 意 力 机 制 ( NAM) 双 水 平 路 由 注 意 视 觉 变 压 器
( BiFormer) |
| DOI： |
| 分类号: |
| 基金项目: |
|
| A Method for Extracting Buildings from Remote Sensing Images Based on Lightweight NDFEDet-SOLOv2 |
|
WANG Qiang GUO Laigong CHENG Weitao
|
|
School of Electrical and Information Engineering Anhui University of Science & Technology Anhui Huainan 232001
China
|
| Abstract: |
| In the fields of cadastral surveying and disaster management automatic extraction of building
contours and positions is crucial. To address the problem of inaccurate segmentation accuracy of buildings in high-
resolution remote sensing images due to environmental factors an improved lightweight SOLOv2 instance segmentation
model called NDFEDet-SOLOv2 was proposed. Methods This model adopted a lightweight EfficientDet network with a
bidirectional feature pyramid network BiFPN feature fusion method. The EfficientNet in the backbone network was
upgraded to EfficientNetv2 and the three-layer MBConv module SE attention in EfficientNetv2 was replaced with a
lightweight normalized attention mechanism NAM containing DropBlock regularization forming the NAD-MBConv
module. The feature fusion part of BiFPN incorporated a bi-level routing attention visual transformer BiFormer into each
feature layer at its tail end to form a bi-directional horizontally routing attention feature pyramid network structure Bi-
FPN-Former which focused on the contour information of small buildings and achieved higher-level feature fusion. Results Compared with traditional lightweight SOLOv2 instance segmentation algorithms the NDFEDet-SOLOv2 model
has improved average accuracies by 3. 9% 3. 7% and 2. 5% for mAP mAP50 and mAP75 respectively and improved
detection frame rate FPS by 4. 7 frames / s. Conclusion The lightweight NDFEDet-SOLOV2 instance segmentation
algorithm model eliminates image distortion of building edges and corners and can accurately extract the basic contours of
buildings in remote sensing images even in complex and uneven geographical environments. This provides a theoretical
reference for the update of urban layouts and the detection of building changes. |
| Key words: high-resolution remote sensing images instance segmentation EfficientDet normalized attention mechanism
NAM bi-level routing attention visual transformer BiFormer |
