EUVIMEDEUVIMED entdecken
Open Access

Enhancing Tactile Paving Segmentation via Multidimensional Perception Networks

Ao Zhang, Haiyan Zhang, Yemeng Zhu, Yinuo Guo, Zhiwen Zhuang, Tingmei Ma, Yuchen Cai, Luyue Liu

Annals of the New York Academy of Sciences · 2026

DOI: 10.1111/nyas.70298 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Annals of the New York Academy of Sciences, 2026. Verfasst von Ao Zhang und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer and Software Engineering, Huaiyin Institute of Technology, Huai'an, Jiangsu, China. Hauptthemen laut MEDLINE: Touch Perception, Touch, Classification Algorithms, Construction Materials. Die Arbeit nennt 42 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Enterprise Collaboration Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Tactile paving is essential for ensuring the safe and independent travel of visually impaired individuals in urban environments. However, existing segmentation models often fail to generalize well across diverse scenarios and rely heavily on color information, neglecting the physical structure of tactile patterns. This article introduces EGM‐Unet, a novel segmentation method that incorporates an edge‐aware multi‐scale attentional fusion block (EMAF‐Block) to enhance edge feature extraction, a recursive gated attention (RGA) mechanism to focus on critical regions, and a multidimensional cooperative aggregation attention (MCAA) module to refine texture features. Additionally, we utilize CLIPSeg with a correlative self‐attention (CSA) mechanism to improve generalization. Our experiments demonstrate that EGM‐Unet outperforms standard U‐Net and other mainstream models, achieving a mean IoU of 93.73%, an IoU of 89.09%, and an accuracy of 98.56%. These results highlight the robustness of our method in segmenting tactile paving regions across various materials and lighting conditions, providing a solid foundation for high‐precision tactile paving perception.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Touch PerceptionTouchClassification AlgorithmsConstruction MaterialsHumansAttentionArtificial IntelligenceConvolutional Neural NetworksImage Processing, Computer-AssistedPersons with Visual Disabilities

Stichwörter der Autor:innen

U‐Netattention mechanismmulti‐scale feature fusiontactile paving segmentation

Autor:innen und Institutionen

  • School of Computer and Software Engineering, Huaiyin Institute of Technology, Huai'an, Jiangsu, China.

Förderung

  • Enterprise Collaboration Project

Zitieren

APA

Zhang, A., Zhang, H., Zhu, Y., Guo, Y., Zhuang, Z., Ma, T., Cai, Y., & Liu, L. (2026). Enhancing Tactile Paving Segmentation via Multidimensional Perception Networks. Annals of the New York Academy of Sciences. https://doi.org/10.1111/nyas.70298

Vancouver

Zhang A, Zhang H, Zhu Y, Guo Y, Zhuang Z, Ma T, et al. Enhancing Tactile Paving Segmentation via Multidimensional Perception Networks. Annals of the New York Academy of Sciences. 2026. doi:10.1111/nyas.70298

BibTeX
@article{Zhang2026,
  author = {Zhang, Ao and Zhang, Haiyan and Zhu, Yemeng and Guo, Yinuo and Zhuang, Zhiwen and Ma, Tingmei and Cai, Yuchen and Liu, Luyue},
  title = {Enhancing Tactile Paving Segmentation via Multidimensional Perception Networks},
  journal = {Annals of the New York Academy of Sciences},
  year = {2026},
  doi = {10.1111/nyas.70298}
}
Kostenlos starten →