GSM-CrackFormer:基于高斯尺度混合模型的多应用场景裂缝检测方法  

GSM-CrackFormer:Crack Detection Method in Multi-Application Scenarios Based on Gaussian Scale Mixed Model

作　　者：黄廷辉 李升典 Huang Tinghui;Li Shengdian(College of Computer and Information Security,Guilin University of Electronic Technology,Guilin 541000)

机构地区：[1]桂林电子科技大学计算机与信息安全学院,桂林541000

出　　处：《计算机辅助设计与图形学学报》2024年第12期2029-2039,共11页Journal of Computer-Aided Design & Computer Graphics

基　　金：广西自然科学基金面上项目(2019GXNSFAA245053);广西科技重大专项(AA19254016);桂林电子科技大学研究生教育创新计划(2023YCXS068).

摘　　要：传统图像处理方法或机器学习方法解决裂缝检测问题通常仅适应特定场景.随着场景切换,此类方法的检测精度会受到显著的影响,在多应用场景下缺乏鲁棒性.为了适应多应用场景,在原裂缝检测方法CrackFormer基础上进行出改进,提出一种基于高斯尺度混合模型的检测方法——GSM-CrackFormer.首先通过高斯尺度混合模型构建描述裂缝特征的高斯分布的模块;然后结合门控机制设计信号转换模块,将由分布生成的裂缝特征信息转化为锐化裂缝语义特征的指导信号,通过新颖的上下采样策略进一步平衡模型感受野与其捕获细节特征能力之间的关系;最后调整损失函数,缓解裂缝像素与非裂缝像素之间不平衡的问题.在多样化场景数据集CrackSeg9k上进行训练和评估的实验结果表明,所提方法优于文中所对比的方法,其全局最佳(ODS)指标达到0.784,单图最佳(OIS)指标达到0.785,平均交并比(MIoU)达到0.828.Traditional image processing methods or machine learning approaches for addressing crack detec-tion problems typically cater to specific scenarios.As the scene changes,the detection accuracy of such me-thods is significantly affected,lacking robustness across multiple application scenarios.To adapt to diverse application contexts,improvements have been made upon the original crack detection method,CrackFormer,leading to the proposal of a detection approach based on the Gaussian Scale Mixture(GSM)model,known as GSM-CrackFormer.Firstly,a Gaussian Scale Mixture model is constructed to describe Gaussian distribu-tions related to crack features.Subsequently,a signal transformer is designed,incorporating a gating me-chanism to convert crack feature information generated by the distributions into guiding signals for enhanc-ing semantic features.Additionally,a novel up-down sampling strategy is introduced to further balance the relationship between model receptive fields and its capability to capture detailed features.By adjusting the loss function,the imbalance issue between crack and non-crack pixels is effectively mitigated.Finally,we conduct experiments on the CrackSeg9k dataset to evaluate the performance of our proposed method.Our experimental results demonstrate that GSM-CrackFormer outperforms the methods compared in the paper,achieving a global best(ODS)of 0.784,a monograph best(OIS)of 0.785,and a mean intersection ratio(MIoU)of 0.828.

关 键 词：裂缝检测 图像处理方法 高斯尺度混合模型 裂缝检测场景 上下采样策略 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]