EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Smartphone-Based Non-Destructive Multimodal Deep Learning Approach Using pH-Sensitive Pitaya Peel Films for Real-Time Fish Freshness Detection

Yixuan Pan, Yujie Wang, Yuzhe Zhou iD, Jiacheng Zhou, Manxi Chen, Dongling Liu, Feier Li, Can Liu iD, Mingwan Zeng, Dongjing Jiang, Xiangyang Yuan iD, Hejun Wu iD

Foods · 19.05.2025 · 14 (10) · 1805

DOI: 10.3390/foods14101805 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Foods, online erschienen am 19.05.2025. Verfasst von Yixuan Pan und 11 weiteren Autor:innen. Beteiligte Einrichtung: College of Science, Sichuan Agricultural University, No.46, Xin Kang Road, Ya'an 625014, China. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Natural Science Foundation of Sichuan Province, the Sichuan Province Science and Technology Support Program, the Sichuan Tea Innovation Team.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The detection of fish freshness is crucial for ensuring food safety. This study addresses the limitations of traditional detection methods, which rely on laboratory equipment and complex procedures, by proposing a smartphone-based detection method, termed FreshFusionNet, that utilizes a pitaya peel pH intelligent indicator film in conjunction with multimodal deep learning. The pitaya peel indicator film, prepared using high-pressure homogenization technology, demonstrates a significant color change from dark red to yellow in response to the volatile alkaline substances released during fish spoilage. To construct a multimodal dataset, 3600 images of the indicator film were captured using a smartphone under various conditions (natural light and indoor light) and from multiple angles (0° to 120°), while simultaneously recording pH values, total volatile basic nitrogen (TVB-N), and total viable count (TVC) data. Based on the lightweight MobileNetV2 network, a Multi-scale Dilated Fusion Attention module (MDFA) was designed to enhance the robustness of color feature extraction. A Temporal Convolutional Network (TCN) was then used to model dynamic patterns in chemical indicators across spoilage stages, combined with a Context-Aware Gated Fusion (CAG-Fusion) mechanism to adaptively integrate image and chemical temporal features. Experimental results indicate that the overall classification accuracy of FreshFusionNet reaches 99.61%, with a single inference time of only 142 ± 40 milliseconds (tested on Xiaomi 14). This method eliminates the need for professional equipment and enables real-time, non-destructive detection of fish spoilage through smartphones, providing consumers and the food supply chain with a low-cost, portable quality-monitoring tool, thereby promoting the intelligent and universal development of food safety detection technology.

Schlagwörter

Stichwörter der Autor:innen

fish freshness detectionfood safetymultimodal deep learningpH-sensitive filmreal-time monitoringsmartphone-based sensing

Autor:innen und Institutionen

  • College of Science, Sichuan Agricultural University, No.46, Xin Kang Road, Ya'an 625014, China.
  • College of Information Engineering, Sichuan Agricultural University, No.46, Xin Kang Road, Ya'an 625014, China.
  • College of Management, Sichuan Agricultural University, No.211, Hui Min Road, Chengdu 611130, China.
  • College of Food Science, Sichuan Agricultural University, No.46, Xin Kang Road, Ya'an 625014, China.

ORCID-Profile: Yuzhe Zhou, Can Liu, Xiangyang Yuan, Hejun Wu

Förderung

  • the Natural Science Foundation of Sichuan Province
  • the Sichuan Province Science and Technology Support Program
  • the Sichuan Tea Innovation Team
  • the National Undergraduate Training Program on Innovation and Entrepreneurship
  • the Provincial Undergraduate Training Program on Innovation and Entrepreneurship
  • the National/Provincial Undergraduate Training Program on Innovation and Entrepreneurship

Zitieren

APA

Pan, Y., Wang, Y., Zhou, Y., Zhou, J., Chen, M., Liu, D., Li, F., Liu, C., Zeng, M., Jiang, D., Yuan, X., & Wu, H. (2025). A Smartphone-Based Non-Destructive Multimodal Deep Learning Approach Using pH-Sensitive Pitaya Peel Films for Real-Time Fish Freshness Detection. Foods, 14(10), 1805. https://doi.org/10.3390/foods14101805

Vancouver

Pan Y, Wang Y, Zhou Y, Zhou J, Chen M, Liu D, et al. A Smartphone-Based Non-Destructive Multimodal Deep Learning Approach Using pH-Sensitive Pitaya Peel Films for Real-Time Fish Freshness Detection. Foods. 2025;14(10):1805. doi:10.3390/foods14101805

BibTeX
@article{Pan2025,
  author = {Pan, Yixuan and Wang, Yujie and Zhou, Yuzhe and Zhou, Jiacheng and Chen, Manxi and Liu, Dongling and Li, Feier and Liu, Can and Zeng, Mingwan and Jiang, Dongjing and Yuan, Xiangyang and Wu, Hejun},
  title = {A Smartphone-Based Non-Destructive Multimodal Deep Learning Approach Using pH-Sensitive Pitaya Peel Films for Real-Time Fish Freshness Detection},
  journal = {Foods},
  year = {2025},
  volume = {14},
  number = {10},
  pages = {1805},
  doi = {10.3390/foods14101805}
}
Kostenlos starten →