混凝土构件表面裂缝分割的改进SegFormer方法
Improved SegFormer Model for Surface Crack Segmentation in Concrete Components
混凝土结构常常会产生表面裂缝,及时对这些表面裂缝进行检测对结构状况评估有重要意义。本文提出了一种改进SegFormer的方法,用于表面裂缝的分割。该方法利用SegFormer模型的网络架构和创新模块(CA),在解码器部分加入了坐标注意力机制模块,其中创新性地引入了沿对角线方向池化,显著提升了模型在像素级预测时对细长形状空间关系的理解能力。通过对公共数据集Concrete3k的实验结果进行对比,改进SegFormer模型展现出最高的IoU值,表明其在裂缝分割任务中表现最佳,在参数量较低的情况下,改进SegFormer模型仍能保持高性能。
Concrete structures commonly exhibit visible surface cracks and detecting these surface cracks is of great significance for structural condition assessment. This paper presents an improved attention mechanism mode for surface crack segmentation. The method leverages the network framework and innovative module (CA) of the improved attention mechanism mode, adds a coordinate attention module to the decoder, and by introducing pooling along the diagonal direction significantly enhances the model's understanding of elongated spatial relationships in pixel-level predictions. Comparative experiments on the public dataset Concrete3k demonstrate that the improved attention mechanism mode achieves the highest IoU value, indicating superior performance in crack segmentation tasks. Additionally, the improved attention mechanism mode maintains high performance with a relatively low parameter count.
| [1] |
孙威.利用压电陶瓷的智能混凝土结构健康监测技术[D].大连:大连理工大学,2010. |
| [2] |
SUN Wei.Health monitoring technology for smart concrete structures using piezoelectric ceramic[D].Dalian:Dalian University of Technology,2010.(in Chinese) |
| [3] |
贺拴海,宋一凡,赵小星,钢筋混凝土梁式结构裂缝特征与损伤评估方法试验研究[J].土木工程学报,2003(2):6-9. |
| [4] |
HE Shuanhai,SONG Yifan,ZHAO Xiaoxing,et al.Experimental study on crack characteristics and damage evaluation for RC beams[J].China Civil Engineering Journal,2003(2):6-9.(in Chinese) |
| [5] |
SUN Hongshuo,SONG Li,YU Zhiwu.Assessing the fatigue damage of concrete structures using automatically classified crack severity level information[J].International Journal of Fatigue,2024,179:108071. |
| [6] |
马亚飞,孙文康,何羽,基于DC-Unet的混凝土桥梁表观裂缝识别方法[J].长安大学学报(自然科学版),2024,44(3):66-75. |
| [7] |
MA Yafei,SUN Wenkang,HE Yu,et al.Surface crack identification method of concrete bridge based on DC-Unet[J].Journal of Chang'an University (Natural Science Edition),2024,44(3):66-75.(in Chinese) |
| [8] |
ZHANG Ruiyang,MENG Libo,MAO Zhu,et al.Spatiotemporal deep learning for bridge response forecasting[J].Journal of Structural Engineering,2021,147(6):04021070. |
| [9] |
NIKHILANAND A,SRIPARNA S.Multi-modal advanced deep learning architectures for breast cancer survival prediction[J].Knowledge-Based Systems,2021,221:106965. |
| [10] |
WEI Yiru,ZHU Zhiliang,YU Hai,et al.A real-time threat image projection (TIP) model based on deep learning for X-ray baggage inspection[J].Physics Letters A,2021,400:127306. |
| [11] |
徐国整,廖晨聪,陈锦剑,基于HU-ResNet的混凝土表观裂缝信息提取[J].计算机工程,2020,46(11):279-285. |
| [12] |
XU Guozheng,LIAO Chencong,CHEN Jinjian,et al. Extraction of apparent crack information of concrete based on HU-ResNet[J].Computer Engineering,2020,46(11):279-285.(in Chinese) |
| [13] |
CHEN B,ZHANG H,WANG G,et al.Automatic concrete infrastructure crack semantic segmentation using deep learning[J].Automation in Construction,2023,152:104950. |
| [14] |
XIANG C,GUO J,CAO R,et al.A Crack-segmentation algorithm fusing transformers and convolutional neural networks for complex detection scenarios[J].Automation in Construction,2023,152:104894. |
| [15] |
张俊杰.基于深度学习的隧道表观病害智能识别研究[D].长沙:中南大学,2023. |
| [16] |
ZHANG Junjie.Intelligent identification of highway tunnel surface defects based on deep learning[D].Changsha:Central South University,2023.(in Chinese) |
| [17] |
张明星,徐健,刘秀平,改进U-Net的路面裂缝检测方法[J].计算机工程与应用,2024,60(24):306-313. |
| [18] |
ZHANG Mingxing,XU Jian,LIU Xiuping et al.Improved U-Net pavement crack detection method[J].Computer Engineering and Applications,2024,60(24):306-313.(in Chinese) |
| [19] |
王琳琳,孟良,卜博雅,改进DeepLabv3+模型的混凝土坝表观裂缝特征提取方法[J].东南大学学报(自然科学版),2024,54(4):929-936. |
| [20] |
WANG Linlin,MENG Liang,BU Boya,et al.Concrete dam apparent crack characteristic extraction based on improved DeepLabv3+ model[J].Journal of Southeast University (Natural Science Edition),2024,54(4):929-936.(in Chinese) |
| [21] |
ELMESSERY M W,MAKLAKOV V D,MESSERY E M T,et al.Semantic segmentation of microbial alterations based on SegFormer[J].Frontiers in Plant Science,2024,15:1352935. |
| [22] |
LI Meilin,RUI Jie,YANG Songkun,et al.Method of building detection in optical remote sensing images based on SegFormer [J].Sensors,2023,23(3):1258-1258. |
| [23] |
陈守存.基于Transformer和对比学习的脑肿瘤图像分割方法研究[D].济南:济南大学,2024. |
| [24] |
CHEN Shoucun.Research on brain tumor image segmentation based on transformer and contrastive learning[D].Jinan:University of Jinan,2024.(in Chinese) |
| [25] |
ZHENG Qingping,ZHENG Ling,et al.Transformer-based hierarchical dynamic decoders for salient object detection[J].Knowledge-Based Systems,2023,282:111075. |
| [26] |
李海丰,范天啸,黄睿,联合Self-attention与Axial-attention的机场跑道裂缝分割[J].郑州大学学报(理学版),2023,55(4):30-38. |
| [27] |
LI Haifeng,FAN Tianxiao,HUANG Rui,et al.Combining Self-attention and axial-attention for airport runway crack segmentation[J].Journal of Zhengzhou University (Natural Science Edition),2023,55(4):30-38.(in Chinese) |
| [28] |
王张钰.基于窗口自注意机制网络的显著性目标检测[D].石家庄:河北师范大学,2024. |
| [29] |
WANG Zhangyu.Salient object detection network based on self-attention with windows[D].Shijiazhuang:Hebei Normal University,2024.(in Chinese) |
| [30] |
LI Y,MA R,LIU H,et al.Real-time high-resolution neural network with semantic guidance for crack segmentation[J].Automation in Construction,2023,156:105112. |
国家自然科学基金(52378540)
/
| 〈 |
|
〉 |