Vectorization of raster mechanical drawings on the base of ternary segmentation and soft computing

Kasimov, Denis R.; Kuchuganov, Aleksandr V.; Kuchuganov, Valeriy N.; Oskolkov, P. P.

doi:10.1134/s0361768817060056

Cited by 4 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To find the overall performance for a vectorisation method, researchers used the unweighted mean of the VRI scores. To the best of our knowledge, this seems to be the case in all of the research publications (Wang et al, 2010;Bonnici and Camilleri, 2013;Wu et al, 2013;Zhang et al, 2015;Kasimov et al, 2017;Bonnici et al, 2019;Alwan et al, 2019) that use VRI as the preferred performance criterion as well as papers published by the authors of VRI (Liu et al, 2001). However, using the unweighted mean is only accurate if all the images (ground truth images) have the same number of primitives (graphical entities) which is rarely the case.…”

Section: Introductionmentioning

confidence: 91%

A Modified Vector Recovery Index

Al-Khaffaf

2020

SJUOZ

View full text Add to dashboard Cite

In this paper, we show that averaging of the Vector Recovery Index (VRI) score for a test involving many images is not accurate and leads to bias. We demonstrate that the higher the difference in primitive count between the data files in an experiment, the higher the bias in calculating the VRI. Normalizing VRI scores is proposed to remove the bias and to get VRI scores that precisely reflects the performance based on images under scrutiny. Empirical performance evaluation on three datasets from the arc segmentation contests attached to International Workshops on Graphics Recognition 2005, 2009, and 2011 shows that the proposed normalization score provides accurate and realistic performance results than the unweighted average of VRI scores. The results based on the modified VRI score show that the vectorisation methods have lower performance than was usually thought.

show abstract