1
Alexander V. Orlov
Kostroma State University
Evgeny L. Pashin
Kostroma State Argicultural Academy
MEASURING THICKNESS OF FLAX FIBRES IN A CLUSTER USING COMPUTER VISION
Orlov A. V., Pashin E. L. Measuring thickness of flax fibres in a cluster using computer vision. Technologies & Quality. 2024. No 1(63). P. 5–11. (In Russ.) https: doi10.34216/2587-6147-2024-1-63-5-11.
DOI: https://doi.org/10.34216/2587-6147-2024-1-63-5-11
УДК: 677.017.4:620.171.3
EDN: GXCFVT
Publish date: 2024-03-06
Annotation: Authors analyse a method of indirect measurement of flax fibre linear density using its thickness, based on computer vision approach and “distance transform” algorithm in particular. A number of flaws of this method are identified. The main reason for the discrepancy is the effect of crossed fibres within the analysed sample. The secondary reason is the specific shape of fibre’s topological skeletonnear its tips. Based on these observations, a software model of the problem has been created. The software produces a simplified image of crossed fibres with specified properties, and measures this image using the above method. The effect of various factors on the discrepancy in the distribution of thickness values appears to match well with theoretical analysis. Additionally, a number of ways to improve the software model are outlined, related to how typicallyfibres are oriented within the analysed sample.
Keywords: computer vision, linear density, bast fibre, distance transform, topological skeleton, crossing, software model
Literature list: 1. Orlov A. V., Pashin E. L. Developing abast fiber linear density calculation algorithm based on computer vision. Izvestiya vysshih uchebnyh zavedenij. Seriya Teknologiya Tekstil’noi Promyshlennosti [Proceedings of Higher Educational Institutions. Series Textile Industry Technology]. 2015;5:65–68. (In Russ.) 2. Orlov A. V., Pashin E. L. Preparation of digital images of a bast fiber sample for optical measurement of its geometrical properties. Tekhnologii i kachestvo [Technology & Quality]. 2018;1:43–47. (In Russ.) 3. Orlov A. V., Pashin E. L. Lighting conditions required for estimation of bast fibers’ thickness using computer vision. Tekhnologii i kachestvo [Technology & Quality]. 2019;1:21–25. (In Russ.) 4. Pashin E. L., Orlov A. V. A method of measuring thickness of bast fiber. Russian Federation patent no. 2779715. Published 19.09.2022, issue 26. (In Russ.) 5. Serra J. Image Analysis and Mathematical Morphology. Orlando, Academic Press, 1982. 610 p. 6. Gibson S. F., Perry R. N., Rockwood A. P., Jones T. R. Adaptively sampled distance fields: a general representation of shape for computer graphics. Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques, SIGGRAPH 2000. New Orleans, Association for Computing Machinery, 2000, pp. 249–254. 7. Ramdas A., Garcia N., Cuturi M. On Wasserstein Two Sample Testing and Related Families of Nonparametric Tests. Entropy. 2017;19(2). URL: https://www.mdpi.com/1099-4300/19/2/47 (accessed 19.11.2023). 8. Dobrushin R. L. Defining a random variable system using conditional districutions. Teoriya veroyatnostej i ee primeneniya [Probability theory & its applications]. 1970;15(3):469–497. (In Russ.)
Author's info: Alexander V. Orlov, Kostroma State University, Kostroma, Russia, aorlov@list.ru, https://orcid.org/0000-0002-4995-3393
Co-author's info: Evgeny L. Pashin, Kostroma State Argicultural Academy, Kostroma, Russia, evgpashin@yandex.ru, https://orcid.org/0000-0002-5871-874Х