用深度学习提升激光扫描仪精度，让便宜设备接近高端效果
Enhancing 3D point accuracy of laser scanner through multi-stage convolutional neural network for applications in construction
- 用多阶段卷积网络建模扫描误差分布，结合几何处理实现精准修正
- 在粗糙室内场景中，均方误差降低超70%，信噪比提升约6分贝
- 适合建筑测绘、旧房改造等需要高精度点云的低成本应用
本文提出一种基于多阶段卷积神经网络（MSCNN）的集成方法，用于降低粗糙室内环境中激光扫描仪（LS）的3D点云定位不确定性，为高精度几何建模与改造提供更准确的空间测量。由于设备限制和环境因素，高端扫描仪（HAS）与低端扫描仪（LAS）存在位置误差。本方法通过在相同环境下配对使用HAS与对应LAS的测量数据，量化特定误差模式。基于测量偏差与其空间分布的统计关系，构建融合传统几何处理与针对性神经网络优化的校正框架。该方法将系统性误差量化转化为监督学习问题，在保留关键几何特征的同时实现精确修正。实验结果表明，在自建粗糙室内场景数据集上，均方误差（MSE）下降超过70%，峰值信噪比（PSNR）提升约6分贝。该方法使低端设备无需硬件升级即可达到接近高端设备的测量不确定性水平。
原文摘要 · Abstract (English)
We propose a multi-stage convolutional neural network (MSCNN) based integrated method for reducing uncertainty of 3D point accuracy of lasar scanner (LS) in rough indoor rooms, providing more accurate spatial measurements for high-precision geometric model creation and renovation. Due to different equipment limitations and environmental factors, high-end and low-end LS have positional errors. Our approach pairs high-accuracy scanners (HAS) as references with corresponding low-accuracy scanners (LAS) of measurements in identical environments to quantify specific error patterns. By establishing a statistical relationship between measurement discrepancies and their spatial distribution, we develop a correction framework that combines traditional geometric processing with targeted neural network refinement. This method transforms the quantification of systematic errors into a supervised learning problem, allowing precise correction while preserving critical geometric features. Experimental results in our rough indoor rooms dataset show significant improvements in measurement accuracy, with mean square error (MSE) reductions exceeding 70% and peak signal-to-noise ratio (PSNR) improvements of approximately 6 decibels. This approach enables low-end devices to achieve measurement uncertainty levels approaching those of high-end devices without hardware modifications.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。