Journal article Open Access

# A novel approach for defect detection on vessel structures using saliency-related features

Francisco Bonnin-Pascual; Alberto Ortiz

### Citation Style Language JSON Export

{
"publisher": "Zenodo",
"DOI": "10.5281/zenodo.4408152",
"container_title": "Ocean Engineering",
"title": "A novel approach for defect detection on vessel structures using saliency-related features",
"issued": {
"date-parts": [
[
2018,
2,
1
]
]
},
"abstract": "<p>Seagoing vessels have to undergo regular visual inspections in order to detect defects such as cracks and corrosion&nbsp;before they result into catastrophic consequences. These inspections are currently performed manually by ship&nbsp;surveyors at a great cost, so that any level of assistance during the inspection process by means of e.g. a fleet of&nbsp;robots capable of defect detection would significatively decrease the inspection cost. In this paper, we describe a&nbsp;novel framework for visually detecting the aforementioned defects. This framework is generic and flexible in the&nbsp;sense that it can be easily configured to compute the features that perform better for the inspection at hand.&nbsp;Making use of this framework and inspired by the idea of conspicuity, this work considers contrast and symmetry&nbsp;as features for detecting defects and shows their usefulness for the case of vessels. Three different combination&nbsp;operators are additionally tested in order to merge the information provided by these features and improve the detection performance. Experimental results for different configurations of the detection framework show better&nbsp;classification rates than state of the art methods and prove its usability for images collected by a micro-aerial&nbsp;robotic platform intended for visual inspection.</p>",
"author": [
{
"family": "Francisco Bonnin-Pascual"
},
{
"family": "Alberto Ortiz"
}
],
"page": "397-408",
"volume": "149",
"note": "This is a preprint version of publication with DOI: https://doi.org/10.1016/j.oceaneng.2017.08.024. This work is partially supported by FEDER funding, by project nr. AAEE50/2015 (Direccio General d'Innovacio i Recerca, Govern de les Illes Balears).",
"type": "article-journal",
"issue": "February",
"id": "4408152"
}
45
85
views