Improved quality assessment of colour surfaces for additive manufacturing based on image entropy

Abstract

A reliable automatic visual quality assessment of 3D-printed surfaces is one of the key issues related to computer and machine vision in the Industry 4.0 era. The colour-independent method based on image entropy proposed in the paper makes it possible to detect and identify some typical problems visible on the surfaces of objects obtained by additive manufacturing. Depending on the quality factor, some of such 3D printing failures may be corrected during the printing process or the operation can be aborted to save time and filament. Since the surface quality of 3D-printed objects may be related to some mechanical or physical properties of obtained objects, its fast and reliable evaluation may also be helpful during the quality monitoring procedures. The method presented in the paper utilizes the assumption of the increase of image entropy for irregularly distorted 3D-printed surfaces. Nevertheless, because of the local nature of distortions, the direct application of the global entropy does not lead to satisfactory results of automatic surface quality assessment. Therefore, the extended method, based on the combination of the local image entropy and its variance with additional colour adjustment, is proposed in the paper, leading to the proper classification of 78 samples used during the experimental verification of the proposed approach.

Quality Assessment of 3D Printed Surfaces Using Combined Metrics Based on Mutual Structural Similarity Approach Correlated with Subjective Aesthetic Evaluation

“…Nevertheless, the development of even better-correlated metrics requires additional experiments, as well as the possible extension of the developed dataset using also some additional samples with non-planar surfaces (with the time-consuming acquisition of subjective quality scores for the new samples). Such an extension of the proposed method for the evaluation of non-planar surfaces is possible similarly as for previously investigated entropy-based approaches [4]. Obviously, reasonable results may be expected only for the division into relatively small regions, additionally avoiding the mutual similarity calculations for distant image fragments due to expected different orientation of both compared to local patterns.…”

“…In comparison to previous studies, the obtained results are superior to those presented in the recent conference paper [58], where PLCC = 0.8353 has been obtained for the combination of various approaches, previously applied only for sample classification purposes. However, the methods applied in the aforementioned paper, apart from the use of FSIMc metric, are based on various additional calculations including preprocessing with the Contrast Limited Adaptive Histogram Equalization (CLAHE) algorithm, entropy calculations [3,4], use of the Histogram of Oriented Gradients (HOG) and-more importantly-entropy of the depth maps obtained using the additional 3D scanning.…”

“…The size of such obtained images is 1600 × 1600 pixels, being equivalent to the physical size of the samples equal to 35 mm × 35 mm. Their thickness is about 4 mm and the height of layers varies from 0.3 to 0.35 mm depending on the individual printer and the size of the nozzle [4]. The additional depth maps, however not utilized in this paper, have been acquired using the ATOS 3D scanner manufactured by the GOM company (more information may be found in the paper [3]).…”

“…Some distortions may also be caused by an improper configuration of stepper motors as well as changing filament's delivery speed. Both these types of distortions may be easily forced during the preparation of test samples, which have been used for the development of the database used for verification of the proposed methods also in some earlier papers [3,4].…”

Análise bibliométrica da produção científica da indústria 4.0 e sua relação com a indústria calçadista

“…Fonte: Kang (2016); Wang (2016); Liu e Xun (2017); Frank et al (2019); Okarma (2020); Pérez, L et al, (2020); Ibrahim, A et al (2019); Vaidya, 2018Vaidya, (2020…”