Roughness prediction of laser cut edges by image processing and artificial neural networks

García, Alberto Tomás; Levichev, Nikita; Vorkov, Vitalii; Rodrigues, Gonçalo Costa; Cattrysse, Dirk; Duflou, Joost R.

doi:10.1016/j.promfg.2021.07.040

Cited by 7 publications

(3 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…New techniques combining image processing and machine learning have been recently developed to estimate roughness [7,8,10]. But, as discussed in [11], these methods are extremely sensitive to lighting conditions and require a large database.…”

Section: Roughnessmentioning

confidence: 99%

“…However, when roughness measurement aims to define the post-process requirement, other feature parameters (for instance, S10z) can be used to shift the importance towards extreme peaks and valleys. By correlating the cutting process parameters with the areal roughness of the cut edge, roughness prediction could be realized with machine learning algorithms, for instance, by using one of the already proposed techniques [7,8,13]. This approach could be further extended to be used for modeling the surface topography of the cut edge depending on the process parameters [14].…”

Section: Fig 1 An Example Of Roughness Measurements: (A) and (B) Cont...mentioning

confidence: 99%

“…An image-based technique to estimate the roughness of laser cutting samples has been described in [7]. In addition, an algorithm to predict the roughness of cutting edges by using an artificial neural network and applying image processing has been developed in [8]. A technique for quantitative identification of cut edge quality factors using an image analysis algorithm has been presented in [2].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Revisiting image-based quality evaluation of laser cut edges

Kardan

2023

Materials Research Proceedings

View full text Add to dashboard Cite

Abstract. The optimization of laser cutting process parameters relies on productivity and obtained edge quality. Whereas maximizing cutting speed is a default procedure to increase the process performance, quality assessment of a cut edge is a non-trivial task. Both contact-based and image-based approaches can be used to quantify the quality of a cut surface. Since contact-based techniques are time-consuming and typically require expert knowledge, the development of simple and fast image-based approaches could improve the performance of sheet metal workshops. Due to the numerous quality characteristics that have to be considered, a significant challenge remains to establish a versatile approach for image-based quality evaluation. Within this paper, the quality assessment of laser cut edges by means of image processing techniques is analyzed. Additionally, the potential for employing visual evaluation to assess all quality indicators in a comprehensive measuring strategy is explored. Finally, the role of the presented approaches in shifting toward intelligent manufacturing is briefly discussed.

show abstract