Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Eine Plattform für die Wissenschaft: Bauingenieurwesen, Architektur und Urbanistik
Rail Surface Defect Detection Method Based on Deep Learning Method with 3D Range Image
In the methods of using images for detecting surface defects of rails, the interaction such as light, stains, and water stains will cause false alarms. This paper proposals a method to detect surface defects of rails using 3D range line scan cameras combined with deep learning. By using the 3D range camera to acquire the information and 2D image information, and optimizing The original internet neural network structure, combined with the channel attention mechanism, a twin unet & 3D+ neural network model is proposed. First, the required database was established by using the 3D range camera, and then the comparison experts provided that the neural network proposed in this paper can effectually eliminate false alarms caused by light, Stains and water stains compared with other neural networks, and effectually promoted the rail surface. The correct rate of default detection.
Rail Surface Defect Detection Method Based on Deep Learning Method with 3D Range Image
In the methods of using images for detecting surface defects of rails, the interaction such as light, stains, and water stains will cause false alarms. This paper proposals a method to detect surface defects of rails using 3D range line scan cameras combined with deep learning. By using the 3D range camera to acquire the information and 2D image information, and optimizing The original internet neural network structure, combined with the channel attention mechanism, a twin unet & 3D+ neural network model is proposed. First, the required database was established by using the 3D range camera, and then the comparison experts provided that the neural network proposed in this paper can effectually eliminate false alarms caused by light, Stains and water stains compared with other neural networks, and effectually promoted the rail surface. The correct rate of default detection.
Rail Surface Defect Detection Method Based on Deep Learning Method with 3D Range Image
Lecture Notes in Civil Engineering
Yang, Yang (Herausgeber:in) / Ming, Geng (Autor:in) / Zhou, Bo (Autor:in) / Luo, Xiaohua (Autor:in) / Ling, Ren (Autor:in) / Zhou, Mingxiang (Autor:in)
14.02.2023
15 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
| DOAJ | 2024
Deep Learning-Based Automatic Defect Detection Method for Sewer Pipelines
| DOAJ | 2023
Deep Learning-Based Defect Detection for Sustainable Smart Manufacturing
| DOAJ | 2022