Ship structure corrosion detection using advanced image processing, active contour algorithm, and parallel processing

  • Md Mahadi Hasan Imran*
  • , Shahrizan Jamaludin
  • , Mohammad Ilyas Khan
  • , Atta ur Rahman
  • , Ahmad Faisal Mohamad Ayob
  • , Muhamad Zalani Bin Daud
  • , Mohammad Fadhli Bin Ahmad
  • , Abdullah Alqahtani
  • , Sayyid Zainal Abidin Bin Syed Ahmad
  • , Nurafnida Binti Afrizal
  • *Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

3 Scopus citations

Abstract

Corrosion poses significant risks to ship infrastructure, causing safety hazards and financial losses. This study introduces a hybrid corrosion detection algorithm combining advanced image processing and a Chan-Vese active contour model with parallel processing. Proposed method achieves classification accuracy of 98 %, precision of 97 %, and segmentation accuracy of 93.83 %. It delivers a sensitivity of 96 % and specificity of 89 % with an 89.2 % lower cutoff. GPU-based acceleration reduces normalization time by 68.49 %, feature extraction by 81.25 %, and matching by 45.01 %. Therefore, proposed method offers a robust and efficient tool for the maritime industry, significantly enhancing the reliability of ship structure monitoring.

Original languageEnglish
Article number122351
JournalOcean Engineering
Volume341
DOIs
StatePublished - 1 Dec 2025

Keywords

  • Active contour
  • Corrosion detection
  • Image processing
  • Ship structure corrosion

Fingerprint

Dive into the research topics of 'Ship structure corrosion detection using advanced image processing, active contour algorithm, and parallel processing'. Together they form a unique fingerprint.

Cite this