Deep learning-based automated 3D inspection of helical gears using voxelized CAD models and 3D convolutional autoencoders

The automated inspection of complex freeform components, such as helical gears, is a persistent challenge in advanced manufacturing due to their intricate geometries and strict precision requirements. Conventional inspection methods, such as those using coordinate measuring machines or optical techn...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	International journal of advanced manufacturing technology Ročník 141; číslo 7-8; s. 3695 - 3715
Hlavní autoři:	Selloum, Rabia, Ameddah, Hacene, Brioua, Mourad
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	London Springer London 01.12.2025 Springer Nature B.V
Témata:	Accuracy Algorithms Automation CAD CAE) and Design Computer aided design Computer-Aided Engineering (CAD Coordinate measuring machines Datasets Deep learning Defects Engineering Helical gears Industrial and Production Engineering Industrial applications Inspection Inspections Machine learning Manufacturing Measurement techniques Mechanical Engineering Media Management Neural networks Noise sensitivity Noise tolerance Optics Optimization Original Article Perturbation Product reliability Deep learning Helical gear inspection Precision manufacturing Convolutional autoencoder Anomaly detection Non-Destructive testing
ISSN:	0268-3768, 1433-3015
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	The automated inspection of complex freeform components, such as helical gears, is a persistent challenge in advanced manufacturing due to their intricate geometries and strict precision requirements. Conventional inspection methods, such as those using coordinate measuring machines or optical techniques, are often time-consuming and lack adaptability to subtle deviations. Recent deep learning approaches show promise but are typically limited to point-based or scan-to-scan comparisons, which remain sensitive to noise and alignment errors. We propose a voxel-based 3D inspection framework that integrates an XGBoost-guided perturbation model with a 3D convolutional autoencoder (3D CNN-AE). CAD-derived gear models are systematically perturbed with controlled Gaussian deformations to emulate tolerances, defects, and sensor noise, then voxelized for autoencoder training. This enables robust learning of nominal gear geometry distributions. Extensive experiments conducted against PointNet++, a Variational Autoencoder, and a GAN-based reconstruction model demonstrate that our method consistently achieves superior performance across various metrics, including PSNR, SSIM, accuracy, precision, recall, and F1-score. The results highlight the potential of voxel-based learning with data-driven perturbation for scalable and high-accuracy inspection in industrial applications.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	0268-3768 1433-3015
DOI:	10.1007/s00170-025-16892-y