Enhanced deep learning models for automatic fish species identification in underwater imagery

Siri D; Gopikrishna Vellaturi; Shaik Hussain Shaik Ibrahim; Srikanth Molugu; Venkata Subbaiah Desanamukula; Raviteja Kocherla; Ramesh Vatambeti

doi:10.1016/j.heliyon.2024.e35217

Enhanced deep learning models for automatic fish species identification in underwater imagery

Heliyon. 2024 Jul 27;10(15):e35217. doi: 10.1016/j.heliyon.2024.e35217. eCollection 2024 Aug 15.

Authors

Siri D¹, Gopikrishna Vellaturi², Shaik Hussain Shaik Ibrahim³, Srikanth Molugu³, Venkata Subbaiah Desanamukula⁴, Raviteja Kocherla³, Ramesh Vatambeti⁵

Affiliations

¹ Department of CSE, Gokaraju Rangaraju Institute of Engineering and Technology, Hyderabad, India.
² Department of Information Technology, MLR Institute of Technology, Hyderabad, India.
³ Department of Computer Science and Engineering, Malla Reddy University, Hyderabad, 500043, India.
⁴ Department of Computer Science and Engineering, Lakireddy Bali Reddy College of Engineering, India.
⁵ School of Computer Science and Engineering, VIT-AP University, Vijayawada, 522237, India.

Abstract

Underwater cameras are crucial in marine ecology, but their data management needs automatic species identification. This study proposes a two-stage deep learning approach. First, the Unsharp Mask Filter (UMF) preprocesses images. Then, an enhanced region-based fully convolutional network (R-FCN) detects fish using two-order integrals for position-sensitive score maps and precise region of interest (PS-Pr-RoI) pooling for accuracy. The second stage integrates ShuffleNetV2 with the Squeeze and Excitation (SE) module, forming the Improved ShuffleNetV2 model, enhancing classification focus. Hyperparameters are optimized with the Enhanced Northern Goshawk Optimization Algorithm (ENGO). The improved R-FCN model achieves 99.94 % accuracy, 99.58 % precision and recall, and a 99.27 % F-measure on the Fish4knowledge dataset. Similarly, the ENGO-based ShuffleNetV2 model, evaluated on the same dataset, shows 99.93 % accuracy, 99.19 % precision, 98.29 % recall, and a 98.71 % F-measure, highlighting its superior classification accuracy.

Keywords: Northern goshawk optimization; Region-based fully convolutional network; ShuffleNet; Squeeze and excitation; Unsharp mask filter.