首页|基于注意力机制的U-Net叶片缺陷图像分割

基于注意力机制的U-Net叶片缺陷图像分割

扫码查看
为解决风力发电机叶片表面缺陷检测存在分类困难和微小缺陷分割模糊的难题,构建一种基于扩张卷积和卷积注意力模块的改进U-Net语义分割网络.该网络基于网络模型的编码-解码结构,使用可迁移的VGG16 的特征提取层代替U-Net网络的编码部分,在编码-解码之间的跳跃模块加入卷积注意力模块.通过对微小缺陷信息选取加强全局权重,使用扩张卷积增强网络特征,采用VGG16 预训练模型实现迁移学习.开展Focal与Dice结合的混合损失函数验证,对比分析DeeplabV3+、PSPnet、HRNet、U-Net这4 种模型.结果表明:对于叶片缺陷数据集,改进的U-Net网络模型对叶片缺陷的分类和分割任务具有更高的精度,均交并比、均像素精度和召回率等指标值分别为83.60%、92.84%和88.50%.改进U-Net网络的均交并比值比DeeplabV3+模型高 13.98%,比标准U-Net模型高9.38%,能够提高叶片缺陷检测的灵敏度,有效降低检测结果的误报警率,有助于准确检测风机叶片缺陷.
Blade defect detection of U-Net network based on attention mechanism
To solve the issues of wind turbine blades in terms of classification difficulty and blurry segmentation of small defects in surface defect detection,an improved U-Net semantic segmentation network was proposed based on dilated convolution and convolutional attention modules.Based on the encoding-decoding structure of the network model,a transferable VGG16 feature extraction layer was used to replace the encoding part of the U-Net network.Then,a convolutional attention module was added to the skip module between encoding and decoding.The global weight was enhanced by selecting small defect information.Dilated convolution was used in the decoding section to enhance the network's feature extraction ability,and the pre-trained VGG16 model was used to realize transfer learning.The hybrid loss function of Focal and Dice was validated against the models of DeeplabV3+,Pyramid Scene Parsing Network(PSPnet),High-Resolution Network(HRNet),and U-Net.The results showed that the improved U-Net network had higher prediction accuracy in blade defect classification and segmentation tasks,mean intersection over union,mean pixel accuray,and recall values were 83.60%,92.84%,and 88.50%,respectively.The mean intersection over union simulated by the improved U-Net model was 13.98%and 9.38%higher than that by the DeeplabV3+and standard U-Net model,respectively.Therefore,the proposed model can improve the sensitivity of blade defect detection,effectively reduce false positives of detection results,and provide guidance to wind turbine blade defect detection.

attention mechanismU-Net networkwind turbine blades defectimage segmentationtransfer learningconvolutional block attention module(CBAM)

祁雷、李宁、梁伟、王峥、刘子梁

展开 >

中海油能源发展股份有限公司 清洁能源分公司 天津 300459

中国石油大学(北京) 安全与海洋工程学院,北京 102249

应急管理部油气生产安全与应急技术重点实验室,北京 102249

注意力机制 U-Net网络 风机叶片缺陷 图像分割 语义分割 迁移学习 卷积块注意力模块(CBAM)

中海油重大科技项目

GD2021ZCAF0021

2024

中国安全科学学报
中国职业安全健康协会

中国安全科学学报

CSTPCD北大核心
影响因子:1.548
ISSN:1003-3033
年,卷(期):2024.34(5)