RGB-based visual encoding of vibration data for gearbox fault diagnosis using U-Net segmentation model

İrfan Kiliç, Gülşah Karaduman, Beyda Tasar, Orhan Yaman · 2026

Read RGB-based visual encoding of vibration data for gearbox fault diagnosis using U-Net segmentation model by İrfan Kiliç, Gülşah Karaduman, Beyda Tasar, Orhan Yaman (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0350838.txt.

Description

İrfan Kiliç and co-authors turned gearbox vibration readings into pictures so an image network could hunt for gear faults. Sensors on a test gearbox produced streams of numbers, hard for vision models to read raw. The team summarized each signal by its highest, lowest, and average values and painted those into the red, green, and blue channels of synthetic images, then trained a segmentation network on them.

After only ten training rounds, the model classified faults with accuracy above ninety-nine percent on the benchmark set. Painting numbers as colors let a vision architecture do what it does best, finding patterns in space. Factory engineers chase exactly this: worn gears caught from vibration before they seize. Whether the trick survives noisier real plants is the open test, but the representation itself is the contribution.

Catalog record

CREATORİrfan Kiliç, Gülşah Karaduman, Beyda Tasar, Orhan Yaman
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE37511
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0350838

Browse the full library · About the Tower of Babel