Browsing by Author "Pereira Barella, Bruno"

Now showing 1 - 2 of 2
  • Thumbnail Image
    ItemOpen Access
    Advanced thermal imaging processing and deep learning integration for enhanced defect detection in carbon fiber-reinforced polymer laminates
    (MDPI, 2025-03-25) Garcia Rosa, Renan; Pereira Barella, Bruno; Garcia Vargas, Iago; Tarpani, José Ricardo; Herrmann, Hans-Georg; Fernandes, Henrique
    Carbon fiber-reinforced polymer (CFRP) laminates are widely used in aerospace, automotive, and infrastructure industries due to their high strength-to-weight ratio. However, defect detection in CFRP remains challenging, particularly in low signal-to-noise ratio (SNR) conditions. Conventional segmentation methods often struggle with noise interference and signal variations, leading to reduced detection accuracy. In this study, we evaluate the impact of thermal image preprocessing on improving defect segmentation in CFRP laminates inspected via pulsed thermography. Polynomial approximations and first- and second-order derivatives were applied to refine thermographic signals, enhancing defect visibility and SNR. The U-Net architecture was used to assess segmentation performance on datasets with and without preprocessing. The results demonstrated that preprocessing significantly improved defect detection, achieving an Intersection over Union (IoU) of 95% and an F1-Score of 99%, outperforming approaches without preprocessing. These findings emphasize the importance of preprocessing in enhancing segmentation accuracy and reliability, highlighting its potential for advancing non-destructive testing techniques across various industries.
  • Thumbnail Image
    ItemOpen Access
    Enhancing fault characterisation in composites using infrared thermography: a bee colony optimisation approach with self-organising maps
    (Taylor and Francis, 2024) Pereira Barella, Bruno; Garcia Rosa, Renan; Barbosa de Oliveira, Gina Maira; Fernandes, Henrique
    This work presents an innovative approach aimed at enhancing the characterisation of discontinuities through the processing of thermographic images. The proposed methodology combines self-organising maps (SOM) with bio-inspired parameter optimisation through bee colony optimisation technique. The primary focus is on improving the quality of the fault quantification metric known as the signal-to-noise ratio (SNR). The goal is to achieve a better fault visualisation, ultimately contributing to the advance of thermography as a non-destructive technique. To validate this novel approach, an experiment was conducted using pulsed thermography on a unidirectional carbon laminate piece measuring 33 ✕100 mm. This specimen was intentionally equipped with three artificial delaminations positioned at different depths on specific layers. The results were then compared against conventional approaches such as principal component analysis, partial least-squares regression and polynomial approximation. The findings from this experiment demonstrated the potential of the proposed approach, i.e. the bee colony optimisation coupled with SOM, on the characterisation of discontinuities using infrared thermography data. There was a 15% improvement on the SNR when using the proposed approach over the other tested approaches. This research makes a noteworthy contribution by offering a promising technique for both the detection and characterisation of faults in composite materials.