改进U^(2)-Net的激光条纹中心线高精度提取  

Improved U^(2)-Net for high-precision extraction of laser stripe centerline

作　　者：陈新禹[1] 孙晓雨 孙延鹏 CHEN Xinyu;SUN Xiaoyu;SUN Yanpeng(College of Artificial Intelligence,Shenyang Aerospace University,Shenyang,Liaoning 110000,China;College of Electronic and Information Engineering,Shenyang Aerospace University,Shenyang,Liaoning 110000,China)

机构地区：[1]沈阳航空航天大学人工智能学院,辽宁沈阳110000 [2]沈阳航空航天大学电子信息工程学院,辽宁沈阳110000

出　　处：《光电子.激光》2025年第1期61-68,共8页Journal of Optoelectronics·Laser

基　　金：辽宁省教育厅基础研究项目(JYT2020018);辽宁省自然科学基金(2021-MS-265)资助项目。

摘　　要：针对复杂环境下激光条纹中心线提取算法稳定性差、精度低等问题,提出一种基于改进U^(2)-Net的中心线提取新方法。首先,在U^(2)-Net网络中加入TSA(transformer-self-attention)、TCA(transformer-cross-attention)模块以提高模型的特征提取能力,实现精准像素级分割,有效去除图像中的噪声、毛刺,为后续中心线提取提供高质量的图像源;其次,根据使用场景特点,对传统Steger方法进行改进,完成激光条纹中心线高精度提取;最后,采用信度评价机制对光条中心点进行精度分析。实验结果表明,本文提出的改进U^(2)-Net相较其他主流语义分割网络具有更高的提取精度、更好的抗噪声性能,在此基础上提取的像素中心点的信度值更高,达到传统Steger算法的1.9倍,满足高精度工业测量的需求。Aiming to address the issues of poor stability and low accuracy of laser stripe centerline extraction algorithm in complex environments,a novel centerline extraction method based on improved U^(2)-Net is proposed.Firstly,TSA(transformer-self-attention)and TCA(transformer-cross-attention)modules are added to the U^(2)-Net network to improve the feature extraction ability of the model,achieve accurate pixel-level segmentation,effectively remove noise and glitches in the image,and provide high-quality image sources for subsequent centerline extraction.Secondly,according to the characteristics of the application scenario,the traditional Steger method is improved to complete the high-precision extraction of the centerline of the laser stripe.Finally,the reliability value evaluation mechanism is used to analyze the accuracy of the center point of the light stripe.Experimental results show that compared with other mainstream semantic segmentation networks,the improved U^(2)-Net proposed in this paper has higher extraction accuracy and better anti-noise performance,and the reliability value of the extracted pixel center point on this basis is higher,reaching 1.9 times that of the traditional Steger algorithm,which meets the needs of high-precision industrial measurement.

关 键 词：线结构光 语义分割 光条中心提取 信度评价 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]