Imagery Dataset for Condition Monitoring of Synthetic Fibre Ropes

Published: 8 September 2023| Version 2 | DOI: 10.17632/by9wy6fxsr.2


The dataset comprises images of synthetic fiber rope (SFR) intended for condition monitoring and predicting remaining useful life (RUL). These images were acquired using a Basler acA2000 camera equipped with a Basler C11-5020-12M-P Premium 12-megapixel lens. We have created an extensive dataset containing a total of 3,089 raw images along with their annotations, representing both typical and defective synthetic fiber ropes (SFRs). The images in the dataset were recorded at a frame rate of 165 frames per second (FPS) and had a resolution of 2000 x 1080 pixels. This dataset encompasses a diverse range of potential defect scenarios that can occur during the SFRs' operational lifespan. These scenarios include, but are not limited to normal, and strand core out conditions. Additionally, it consists of image dataset named ‘extra’ depicting the images that need to be neglected in case of generative models. The primary purpose of this dataset is to support computer vision applications, including object detection, classification, and segmentation. These applications aim to identify and analyze defects in SFRs effectively. The availability of this dataset will greatly facilitate the development and assessment of robust defect detection algorithms.



Aalborg Universitet


Image Processing, Non-Destructive Testing, Synthetic Fiber, Automated Segmentation, Detection Technique, Computer Imaging, Damage Classification, Generative Adversarial Network


Energiteknologisk udviklings- og demonstrationsprogram