Online yarn hairiness– Loop & protruding fibers dataset

• Main Author: Pereira, Filipe
• Publication Date: 2024
• Other Authors: Pinto, Leandro, Soares, Filomena, Vasconcelos, Rosa, Machado, José, Carvalho, Vítor
• Format: Article
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/11110/3060; https://doi.org/10.1016/j.dib.2024.110355
• topic: Dataset; Deep learning; Mechatronic system; Yarn defects; Image acquisition; Yarn hairiness; Augmentation

This paper introduces an online dataset focused on detect- ing hairiness in yarn, including loop and protruding fibers. The dataset is designed for use in assessing artificial intelli- gence algorithms. The dataset consists of 684 original images. Through augmentation, this number increases to 1644, with 11,037 annotations derived from videos featuring 56.4tex pur- ple cotton yarn. The videos were captured during the wind- ing and unwinding processes of the purple yarn coil. An im- age acquisition system capable of capturing high-resolution images while the yarn is in motion was used, reaching speeds of up to 4.2 m/s and producing images with a reso- lution of 1.6M pixels. This dataset containing 100m of purple cotton yarn images was recorded and is available for down- load in various formats, including, among others, YOLOv8, YOLOv5, YOLOv7, MT-YOLOv6, COCO JSON, YOLO Darknet, Pascal VOC XML, TFRecord, CreateML JSON. Within an inter- face developed for a designed mechatronic prototype, users can choose to gather images or videos of yarn. Various char- acteristics of the yarn, such us: diameter, linear mass, vol- ume, twist orientation, twist step, number of cables, hairi- ness index, number of loose fibers, thin places, thick places, neps (mass parameters) and U, CV and sH (statistical param- eters) can be obtained. Recently, this online yarn spinning