多线激光光条图像缺陷分割模型研究  

Research on Segmentation Model of Multi-Line Laser Strip Image’s Defects

作　　者：郭晓轩 冯其波[1,3] 冀振燕 郑发家[1] 杨燕燕 GUO Xiao-xuan;FENG Qi-bo;JI Zhen-yan;ZHENG Fa-jia;YANG Yan-yan(School of Physical Science and Engineering,Beijing Jiaotong University,Beijing 100044,China;School of Software,Beijing Jiaotong University,Beijing 100044,China;Dongguan Nannar Technology Co.,Ltd.,Dongguan,Guangdong 523050,China)

机构地区：[1]北京交通大学物理科学与工程学院,北京100044 [2]北京交通大学软件学院,北京100044 [3]东莞市诺丽科技股份有限公司,广东东莞523050

出　　处：《电子学报》2023年第1期172-179,共8页Acta Electronica Sinica

基　　金：国家自然科学基金重点项目(No.51935002);国家自然科学基金面上项目(No.52175493)。

摘　　要：受环境干扰以及反射光影响,室外采集的多线激光光条图像含有光斑和断裂缺陷.为了准确地分割图像缺陷,本文提出了一个轻量的UT(U-shape Target,U代表U型编解码网络结构,T代表靶形视野)分割模型,模型由3×3卷积和靶形卷积堆叠而成.靶形卷积是针对激光光条图像特点提出的多视野卷积模块,模块中四个卷积分支构成靶形卷积视野,能够提取激光光条图像几何结构特征、局部细节特征以及环绕纹理特征.实验表明,UT模型在多线激光光条图像上的缺陷分割精度高于主流分割模型,而且实现了分割精度和参数量的平衡.Influenced by environmental interference and reflected light, multi-line laser strip images collected outdoors contain the defects of flares and fractures. In order to segment the defects accurately, this paper proposes light-weight UT(U-shape Target, U represents a U-shaped encoder-decoder network architecture, and T represents a target-shaped receptive field) segmentation model, which stacks 3 × 3 convolutions and target convolutions. Considering the characteristics of laser strip images, we propose the target convolution, a multiple-receptive-field convolution module. Four convolution branches in this module form a target-shaped convolution receptive field, which can extract the geometric structure features,the local detail features and the surrounding texture features from the laser strip images. Experiments show that the UT model has higher defect segmentation accuracy than mainstream segmentation models, and can achieve the balance between the segmentation accuracy and the number of parameters.

关 键 词：缺陷分割 激光图像 深度学习 轻量级分割模型 多视野卷积 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]