TranSR-Ne RF:Super-resolution neural radiance field for reconstruction and rendering of weak and repetitive texture of aviation damaged functional surface

作     者：Qichun HU Haojun XU Xiaolong WEI Yizhen YIN Weifeng HE Xinmin HAN Caizhi LI Qichun HU;Haojun XU;Xiaolong WEI;Yizhen YIN;Weifeng HE;Xinmin HAN;Caizhi LI

作者机构：National Key Lab of Aerospace Power System and Plasma TechnologyAir Force Engineering UniversityXi’an 710038China iHarbour Academy of Frontier Equipment Institute of Aero-EngineXi’an Jiaotong UniversityXi’an 710049China 

出 版 物：《Chinese Journal of Aeronautics》 (中国航空学报（英文版）)

年 卷 期：2024年第37卷第11期

页      面：447-461页

核心收录：

学科分类：08[工学] 0825[工学-航空宇航科学与技术] 

基　　金：supported by the National Science and Technology Major Project,China(No.J2019-Ⅲ-0009-0053) the National Natural Science Foundation of China(No.12075319) 

主　　题：Functional surface Multi-view reconstruction Neural rendering TranSR-NeRF Image super-resolution Deep learning 

摘      要：In order to reconstruct and render the weak and repetitive texture of the damaged functional surface of aviation,an improved neural radiance field,named TranSR-NeRF,is *** this paper,a data acquisition system was designed and *** acquired images generated initial point clouds through ***,after extracting features from the images through the improved SE-ConvNeXt network,the extracted features were aligned and fused with the initial point cloud to generate high-quality neural point *** ray-tracing and sampling of the neural point cloud,the ResMLP neural network designed in this paper was used to regress the volume density and radiance under a given viewing angle,which introduced spatial coordinate and relative positional *** reconstruction and rendering of arbitrary-scale super-resolution of damaged functional surface is *** this paper,the influence of illumination conditions and background environment on the model performance is also studied through experiments,and the comparison and ablation experiments for the improved methods proposed in this paper is *** experimental results show that the improved model has good ***,the application experiment of object detection task is carried out,and the experimental results show that the model has good practicability.