Strain measurement in tapered optical fiber sensors with deep learning using speckle pattern imaging

Serbetci H., NAVRUZ İ., ARI F.

Optoelectronics and Advanced Materials, Rapid Communications, cilt.19, sa.1-2, ss.17-24, 2025 (SCI-Expanded) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 19 Sayı: 1-2
  • Basım Tarihi: 2025
  • Dergi Adı: Optoelectronics and Advanced Materials, Rapid Communications
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Compendex
  • Sayfa Sayıları: ss.17-24
  • Anahtar Kelimeler: CNN, Deep learning, Optical fiber sensor, Speckle pattern, Strain, Tapered optical fiber
  • Ankara Üniversitesi Adresli: Evet

Özet

In this study, a low-cost and compact tapered optical fiber (TOF) sensor for strain measurement was proposed and experimentally demonstrated. It has been proven that the speckle pattern images recorded with a CCD camera at the fiber output are highly sensitive to the strain applied to the sensor region, and therefore the proposed sensor can measure at a very high resolution of 12.5 με. The effect of sensor geometry on sensor sensitivity was investigated through correlation analysis, and it was shown that sensor sensitivity increased with decreasing waist diameter and increasing waist length of the TOF sensor. In the training of the dataset, real data obtained from experimental studies were used and a strain detection architecture based on convolutional neural networks (CNN) was developed. A high-resolution and high-sensitivity fiber sensor capable of measuring strain with high accuracy has been developed using deep learning network.