Academic Journal
Evaluation of image segmentation approaches for non-destructive detection and quantification of corrosion damage on stonework
| Title: | Evaluation of image segmentation approaches for non-destructive detection and quantification of corrosion damage on stonework |
|---|---|
| Authors: | Μαραβελακη Παγωνα(http://users.isc.tuc.gr/~pmaravelaki), Maravelaki Pagona-Noni(http://users.isc.tuc.gr/~pmaravelaki), Ζερβακης Μιχαηλ(http://users.isc.tuc.gr/~mzervakis), Zervakis Michail(http://users.isc.tuc.gr/~mzervakis), Kapsalas P.() |
| Publisher Information: | Elsevier BV |
| Subject Terms: | Statistical tests, Black crust, Performance evaluation, Stone decay, Image segmentation algorithms |
| Description: | Summarization: This paper approaches the non-destructive analysis of corrosion damage by testing and evaluating several image segmentation schemes for the detection of decay areas. The application test bed for algorithmic evaluation considers stonework surfaces for corrosion damage. Each of the detection approaches handles in a different way the background inhomogeneities. A semi-automated framework for validating the algorithms’ performance is introduced. This framework guarantees reliable and objective estimation of algorithms’ response, while also enabling informed experimental feedback for the design of improved segmentation algorithms. Further to elaborating on the robust points of each segmentation approach, this work also studies the corrosion mechanisms. The latter process involves investigation of the degradation state as reflected by the size of the decay areas and their darkness. The derived assessments closely converge to assessments based on chemical analyses, performed on the same stone surfaces. Presented on: Corrosion Science |
| Document Type: | Article |
| Language: | English |
| DOI: | 10.1016/j.corsci.2007.03.049https://www.sciencedirect.com/science/article/pii/s0010938x07001412 |
| Access URL: | http://purl.tuc.gr/dl/dias/55C4C19B-AC59-42B2-A791-6173B02DB1C2 |
| Accession Number: | edsair.od......4037..a5d491bfa6debc88f35dba22000c8ba8 |
| Database: | OpenAIRE |
Be the first to leave a comment!