Open Access Open Access  Restricted Access Subscription Access

EcoFishID:A CNN-Powered Web Application for Fish Species Recognition and Ecological Profiling

Akshaya P, Anagha G, Dyuthi D, Nishana N, Jency Mariyam John

Abstract


Accurate fish species identification is crucial for marine biodiversity conservation, sustainable fisheries, and ecological research. Traditional methods are manual, time-intensive, and error-prone, necessitating automation. This study introduces EcoFishID, a deep learning-based system leveraging a fine-tuned ResNet-50 for robust feature extraction and classification. Trained on a curated dataset of marine and freshwater species, the model incorporates data augmentation (rotation, contrast enhancement, flipping) to enhance generalization. Integrated into a user-friendly web platform, EcoFishID enables real-time species identification and ecological data retrieval. Performance evaluation demonstrates 99.72% classification accuracy, with high precision and recall in distinguishing morphologically similar species. By automating fish recognition, EcoFishID offers an efficient, scalable tool for ecological research and conservation efforts.

Full Text:

PDF

References


Petrellis, N. (2021). Fish Morphological Feature Recognition Based on Deep Learning Techniques. 2021 10th International Conference on Modern Circuits and Systems Technologies (MOCAST). IEEE. DOI: [10.1109/MOCAST52088.2021.9493407](https://doi.org/10.1109/MOCAST52088.2021.9493407).

Pudaruth, S., Nazurally, N., Appadoo, C., Kishnah, S., & Chady, F. (2021). SuperFish: A Mobile Application for Fish Species Recognition Using Image Processing Techniques and Deep Learning. International Journal of Computing and Digital Systems, 10(1), 1157-1165. DOI: [10.12785/ijcds/1001104](https://dx.doi.org/10.12785/ijcds/1001104).

Zhang, X., Huang, B., Chen, G., Radenkovic, M., & Hou, G. (2023). WildFishNet: Open Set Wild Fish Recognition Deep Neural Network With Fusion Activation Pattern. *IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 16*, 7303-7314. DOI: [10.1109/JSTARS.2023.3299703](https://doi.org/10.1109/JSTARS.2023.3299703).

Mujtaba, D. F., & Mahapatra, N. R. (2021). Fish Species Classification with Data Augmentation. *2021 International Conference on Computational Science and Computational Intelligence (CSCI)*. IEEE. DOI: [10.1109/CSCI54926.2021.00307](https://doi.org/10.1109/CSCI54926.2021.00307).

Shreesha, S., Pai, M. M. M., Verma, U., & Pai, R. M. (2023). Fish Tracking and Continual Behavioral Pattern Clustering Using Novel Sillago Sihama Vid (SSVid). *IEEE Access, 11*, 29400-29414. DOI: [10.1109/ACCESS.2023.3247143](https://doi.org/10.1109/ACCESS.2023.3247143).

Veluswami, S., & Panneerselvam, S. (2022). Multi-species Fish Identification using Hybrid DeepCNN with Refined Squeeze and Excitation Architecture. Aquatic Sciences and Engineering, 37(4), 220-228.

.Huang, H., Zhou, H., Yang, X., Zhang, L., Qi, L., & Zang, A.-Y. (2019). Faster R-CNN for Marine Organisms Detection and Recognition Using Data Augmentation.

Neurocomputing, 337, 372–384.

.Jalal, A., Salman, A., Mian, A., Shortis, M., & Shafait, F. (2020). Fish detection and species classification in underwater environments using deep learning with temporal

information. Ecological Informatics, 57, 101088.

.Jayasundara, D., Ramanayake, L., Senarath, N., & Herath, S. (2020). Multispectral Imaging for Automated Fish Quality Grading. IEEE 15th International Conference on

Industrial and Information Systems (ICIIS), pp. 321–326.

.Ju, Z., & Xue, Y. (2020). Fish species recognition using an improved AlexNet model. Optik, 223, 165499.


Refbacks

  • There are currently no refbacks.