Development and Evaluation of a CNN-Based Real-Time Fish Species Identification System for Fish Sold in Public Markets in the Upper Valley Area of South Cotabato
Keywords:
Fish species classification, Convolutional Neural Network (CNN), Real-time classification, YOLOv8, Consumer protectionAbstract
Seafood mislabeling undermines consumer confidence, food safety, and sustainable
fisheries management, and ordinary consumers often struggle to visually distinguish fish species
sold in public markets. Existing Convolutional Neural Network (CNN)-based fish classification
systems are largely designed for underwater imagery or fisheries monitoring rather than real-time,
consumer-facing use at the point of sale. This study presents the development and evaluation of
a CNN-based real-time fish species identification system for consumers purchasing fish in public
markets in the Upper Valley Area of South Cotabato, Philippines. A ResNet50 model was trained
on a combined dataset of 40 fish species and 22,304 images, integrated with YOLOv8-based
detection and converted to ONNX format for lightweight smartphone deployment, with practical
field application focused on five locally common species: Bangus, Catfish, Tilapia, Horse
Mackerel, and Shrimp. System performance was evaluated using classification accuracy,
precision, recall, F1-score, and end-to-end processing latency. The developed model achieved an
overall classification accuracy of 93.57%, a weighted precision, recall, and F1-score of 0.94, and
completed classification in approximately 65 milliseconds per image, well within the threshold
generally considered acceptable for real-time mobile applications. These findings confirm that
CNN-based image classification can reliably and rapidly identify commercially sold fish species,
supporting its viability as a fast, accurate, and locally relevant tool that strengthens consumer
protection and supports informed purchasing decisions in local seafood markets.
References
Vindigni G, Pulvirenti A, Alaimo S, Monaco C, Spina D, Peri I. Bioinformatics approach to mitigate mislabeling in EU seafood market and protect consumer health. Int J Environ Res Public Health. 2021;18(14):7497.
Ryburn SJ, Ballantine WM, Loncan FM, Manning OG, Alston MA, Steinwand B, et al. Public awareness of seafood mislabeling. PeerJ. 2022;10:e13486.
He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas; 2016. p. 770-8.
Ben Tamou A, Benzinou A, Nasreddine K. Targeted data augmentation and hierarchical classification with deep learning for fish species identification in underwater images. J Imaging. 2022;8(8):214.
Mohammadisabet A, Hasan R, Dattana V, Mahmood S, Hussain S. CNN-based optimization for fish species classification: tackling environmental variability, class imbalance, and real-time constraints. Information. 2025;16(2):154.
Pan SJ, Yang Q. A survey on transfer learning. IEEE Trans Knowl Data Eng. 2010;22(10):1345-59.
Deng J, Dong W, Socher R, Li LJ, Li K, Fei-Fei L. ImageNet: a large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Miami; 2009. p. 248-55.
Allken V, Handegard NO, Rosen S, Schreyeck T, Mahiout T, Malde K. Fish species identification using a convolutional neural network trained on synthetic data. ICES J Mar Sci. 2019;76(1):342-9.
Rauf HT, Lali MIU, Zahoor S, Shah SZH, Rehman AU, Bukhari SAC. Visual features based automated identification of fish species using deep convolutional neural networks. Comput Electron Agric. 2019;167:105075.
Elbatsh K, Sokar I, Ragab S. WikiFish: mobile app for fish species recognition using deep convolutional neural networks. In: Proceedings of the 2021 4th International Conference on Computational Intelligence and Intelligent Systems (CIIS '21). 2022. p. 13-8.
Cusa M, Falcão L, De Jesus J, Biolatti C, Blondeel L, Bracken FSA, et al. Fish out of water: consumers' unfamiliarity with the appearance of commercial fish species. Sustain Sci. 2021;16(4):1313-22.
Priyankan K, Fernando TGI. Mobile application to identify fish species using YOLO and convolutional neural networks. In: Shakya S, Balas VE, Haoxiang W, Baig Z, editors. Proceedings of International Conference on Sustainable Expert Systems. Vol 176. Springer Singapore; 2021. p. 303- 17.
Helmud E, Widodo CE, Nurhayati OD. Fish species classification through deep ensemble learning: multi-scale feature integration and model averaging strategy. Math Model Eng Probl. 2026;13(2):347- 55.
Jocher G, Chaurasia A, Qiu J. Ultralytics YOLOv8 (Version 8.0.0) [Computer software]. 2023. Available from: https://github.com/ultralytics/ultralytics
Agrawal SK. Metrics to evaluate your classification model to take the right decisions [Internet]. Analytics Vidhya; 2021 [cited 2026 Aug 3]. Available from: https://www.analyticsvidhya.com/blog/2021/07/metrics-to-evaluate-your-classification-model-to-t ake-the-right-decisions/
