Investigating Instance Segmentation Performance in Relation to Dataset Composition and Quality
IEEE Access · 2026 · 14 · 110434-110445
DOI: 10.1109/access.2026.3715007 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Martell Bell und 1 weiteren Autor:innen. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: University of Iowa with support from U.S. Department of Education.
Abstract (Originalsprache)
High-mix, low-volume (HMLV) manufacturing faces persistent barriers to deploying deep learning-based perception due to the cost and effort of collecting large, labeled image datasets. This study investigates how training dataset composition, specifically the proportion of real images and the use of synthetic and/or augmented supplementation, relates to instance segmentation performance on real images. Using the Sandcast Images for Foundry Automation (SIFA) dataset, a fixed test set of 108 representative real images was extracted and held constant across all trained models. From the remaining real, augmented, and synthetic images, training pools were generated for each supplementation strategy and ratio of real data. For every condition, 3 random seeds were used to split each pool into 756 training images and 216 validation images. YOLOv8 instance segmentation models initialized with MS COCO weights were evaluated on the same real image test set. Objective image quality metrics were computed for each training subset to characterize dataset structure and relate these properties to model fitness. Under the controlled conditions of this study, introducing a small fraction of real data was associated with substantial performance gains, after which improvements diminished. Datasets combining synthetic and augmented images with limited real subsets generally achieved performance comparable to the all real baseline. Overall, the results suggest effective training datasets benefit from structured, complementary variation that remains anchored to the real image domain, highlighting pathways toward more data efficient model development for HMLV applications.
Förderung
- University of Iowa with support from U.S. Department of Education
Zitieren
APA
Bell, M., & Vitali, R. V. (2026). Investigating Instance Segmentation Performance in Relation to Dataset Composition and Quality. IEEE Access, 14, 110434-110445. https://doi.org/10.1109/access.2026.3715007
Vancouver
Bell M, Vitali RV. Investigating Instance Segmentation Performance in Relation to Dataset Composition and Quality. IEEE Access. 2026;14:110434-110445. doi:10.1109/access.2026.3715007
BibTeX
@article{Bell2026,
author = {Bell, Martell and Vitali, Rachel V.},
title = {Investigating Instance Segmentation Performance in Relation to Dataset Composition and Quality},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {110434--110445},
doi = {10.1109/access.2026.3715007}
}
EUVIMED