| 陈法法,董海飞,何向阳,陈保家.改进U_Net网络的钢结构表面锈蚀图像分割方法[J].电子测量与仪器学报,2024,38(2):49-57 |
| 改进U_Net网络的钢结构表面锈蚀图像分割方法 |
| Improved steel structure surface rust imagesegmentation method for U_Net network |
| |
| DOI： |
| 中文关键词: 锈蚀区域分割 MobilenetV3 U_Net 注意力导向 深度可分离残差卷积 |
| 英文关键词:rust area segmentation MobileNetV3 U_Net attention guided depth separable residual convolution |
| 基金项目:国家自然科学基金( 51975324)、国家大坝安全工程技术研究中心开放基金(CX2022B06)、湖北省教育厅科研项目( B2021036)资助 |
|
|
| Author | Institution |
| Chen Fafa | 1.Hubei Key Laboratory of Hydropower Machinery Design & Maintenance, China Three Gorges University,
Yichang 443002，China;2.National Dam Safety Research Center, Wuhan 430010, China |
| Dong Haifei | Hubei Key Laboratory of Hydropower Machinery Design & Maintenance, China Three Gorges University,
Yichang 443002，China |
| He Xiangyang | National Dam Safety Research Center, Wuhan 430010, China |
| Chen Baojia | Hubei Key Laboratory of Hydropower Machinery Design & Maintenance, China Three Gorges University,
Yichang 443002，China |
|
| 摘要点击次数: 231 |
| 全文下载次数: 333 |
| 中文摘要: |
| 为实现锈蚀图像分割网络模型轻量化,同时消除非单一特征背景和锈液等类似特征背景干扰,本文将U_Net网络模型的编码部分替换为MobilenetV3_Large网络,导入基于ImageNet数据集的MobilenetV3_Large网络预训练权重,将U_Net网络模型解码部分的普通卷积替换为深度可分离残差卷积,并在上采样的过程中添加注意力导向AG模块和Dropout机制。经实验验证表明,本文设计的改进U_Net网络模型在非单一特征背景和锈液等类似特征背景干扰下,具有明显的锈蚀图像分割优势,相比于原U_Net网络模型,模型大小减少了81.18%,浮点计算量减少了98.34%,检测效率提升了3.27倍,即从原来不足6 fps,提升至19 fps。网络模型实现轻量化的同时,网络模型的准确率达95.54%,相比于原U_Net网络模型提升了5.04%。 |
| 英文摘要: |
| In order to lighten the rust image segmentation network model and eliminate the interference of non single feature background and similar feature backgrounds such as rust liquid, this paper replaces the encoded part of the U-Net network model with the MobilenetV3_large network, imports the pre-trained weights of the MobilenetV3_large network based on the ImageNet dataset, and replaces the ordinary convolution of the decoded part of the U-Net network model with a deep separable residual convolution. And add the attention-oriented AG module and the Dropout mechanism in the process of upsampling. Experimental results demonstrate that the improved U-Net network model designed in this paper exhibits significant advantages in rust image segmentation under non-uniform feature background and similar feature background interference such as rust liquids. The model size is reduced by 81.18% compared to the original U-Net network model, resulting in a decrease of floating point calculations by 98.34%. Additionally, the detection efficiency has improved by 3.27 times, increasing from less than 6 frames/s to 19 frames/s. While the network model is lightweight, the accuracy of the network model is 95.54%, which is 5.04% higher than the original U_Net network model. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
