Dual Encoder Ensemble with EfficientNet and VGG for Bird Species Identification

International Journal of Electronics and Communication Engineering

© 2025 by SSRG - IJECE Journal

Volume 12 Issue 10

Year of Publication : 2025

Authors : Sireesha Abotula, Srinivas Gorla, Prasad Reddy PVGD, Dasari Siva Krishna

10.14445/23488549/IJECE-V12I10P101

How to Cite?

Sireesha Abotula, Srinivas Gorla, Prasad Reddy PVGD, Dasari Siva Krishna, "Dual Encoder Ensemble with EfficientNet and VGG for Bird Species Identification," SSRG International Journal of Electronics and Communication Engineering, vol. 12,  no. 10, pp. 1-8, 2025. Crossref, https://doi.org/10.14445/23488549/IJECE-V12I10P101

Abstract:

The identification of bird species is crucial in various fields such as wildlife conservation, ecological research, and biodiversity monitoring. Identifying bird species from images is traditionally labor-intensive and prone to errors, particularly due to the global diversity of avian species. In this work, we propose an Image-Based Bird Species Identification (IMGBSI-NET) system using Deep Learning, which is both highly accurate and innovative. The system integrates two encoders, such as VGG-19 and EfficientNet-B0 pre-trained models. In this approach, a Dual Encoder-based architecture is employed, where various stages of the VGG-19 and EfficientNet models are used to extract features for prediction in bird species identification. In contrast to previous studies, this novel approach enhances the model performance of bird species recognition across a wide range of species. The most appropriate in this framework is EfficientNet, which extracts potent features that can capture fine-grained details that are required to distinguish between species of birds. The suggested method also proves to be compatible with other trained models and various mechanisms. The bird species recognition ability introduced in this study sets a new standard, and its performance is much higher in comparison to state-of-the-art models. The solution is a strong and effective one, and provides the perspectives of future research and practice in computer vision.

Keywords:

EfficientNet, Encoder, Decoder, Bird Species, Features, Deep Learning.

References:

[1] Anne L. Alter, and Karen M. Wang, “An Exploration of Computer Vision Techniques for Bird Species Classification,” stanford.edu, pp. 1-7, 2017.
[Google Scholar] [Publisher Link]
[2] John Atanbori et al., “Classification of Bird Species from Video Using Appearance and Motion Features,” Ecological Informatics, vol. 48, pp. 12-23, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[3] Madhuri Tayal, “Bird Identification by Image Recognition,” Helix, vol. 8, no. 6, pp. 4349-4352, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[4] Jinhai Cai et al., “Sensor Network for the Monitoring of Ecosystem: Bird Species Recognition,” 2007 3rd International Conference on Intelligent Sensors, Sensor Networks and Information, Melbourne, VIC, Australia, pp. 293-298, 2007.
[CrossRef] [Google Scholar] [Publisher Link]
[5] Akash Kumar, and Sourya Dipta Das, “Bird Species Classification using Transfer Learning with Multistage Training,” Workshop on Computer Vision Applications, Hyderabad, India, pp. 28-38, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[6] Ahmad B.A. Hassanat, “Furthest-Pair-Based Binary Search Tree for Speeding Big Data Classification Using K-Nearest Neighbors,” Big Data, vol. 6, no. 3, pp. 171-235, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[7] Baowen Qiao et al., “Bird Species Recognition Based on SVM Classifier and Decision Tree,” 2017 First International Conference on Electronics Instrumentation & Information Systems (EIIS), Harbin, China, pp. 1-4, 2017.
[CrossRef] [Google Scholar] [Publisher Link]
[8] Yo-Ping Huang, and Haobijam Basanta, “Bird Image Retrieval and Recognition Using a Deep Learning Platform,” IEEE Access, vol. 7, pp. 66980-66989, 2019.
[CrossRef] [Google Scholar] [Publisher Link]
[9] Satyam Raj et al., “Image-Based Bird Species Identification using Convolutional Neural Network,” International Journal of Engineering Research & Technology, vol. 9, no. 6, pp. 1-6, 2020.
[CrossRef] [Google Scholar] [Publisher Link]
[10] Vemula Omkarini, and G. Krishna Mohan, “Automated Bird Species Identification Using Neural Networks,” Annals of the Romanian Society for Cell Biology, vol. 25, no. 6, pp. 5402-5407, 2021.
[Google Scholar] [Publisher Link]
[11] S. Dhamodaran et al., “Bird Species Identification from Image,” Mathematical Statistician and Engineering Application, vol. 71, pp. 495-501, 2022.
[Google Scholar]
[12] Aleena Varghese, K. Shyamkrishna, and M. Rajeswari, “Utilization of Deep Learning Technology in Recognizing Bird Species,” AIP Conference Proceedings, vol. 2463, no. 1, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[13] Pureti Anusha, and Kundurthi ManiSai, “Bird Species Classification Using Deep Learning,” 2022 International Conference on Intelligent Controller and Computing for Smart Power (ICICCSP), Hyderabad, India, pp. 1-5, 2022.
[CrossRef] [Google Scholar] [Publisher Link]
[14] Seppo Fagerlund, “Bird Species Recognition Using Support Vector Machines,” EURASIP Journal on Advances in Signal Processing, vol. 2007, pp. 1-8, 2007.
[CrossRef] [Google Scholar] [Publisher Link]
[15] Li Jian, Zhang Lei, and Yan Baoping, “Research and Application of Bird Species Identification Algorithm Based on Image Features,” 2014 International Symposium on Computer, Consumer and Control, Taichung, Taiwan, pp. 139-142, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[16] B. Persia Abishal, and Sujitha Juliet, “Image-Based Bird Species Identification Using Machine Learning,” 2023 9th International Conference on Advanced Computing and Communication Systems FIGICACCS), Coimbatore, India, pp. 1963-1968, 2023.
[CrossRef] [Google Scholar] [Publisher Link]
[17] Sachin Kumar et al., “Enhancing Bird Species Identification Using a Custom CNN Architecture,” 2024 4th International Conference on Technological Advancements in Computational Sciences (ICTACS), Tashkent, Uzbekistan, pp. 1333-1338, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[18] Jashanpreet Kaur et al., “Enhanced Bird Species Identification using ResNet-50: A Deep Learning Framework for High-Performance Classification,” 2024 3rd International Conference on Automation, Computing and Renewable Systems (ICACRS), Pudukkottai, India, pp. 821-826, 2024.
[CrossRef] [Google Scholar] [Publisher Link]
[19] Javier Rodriguez-Juan et al., “Visual Wetlandbirds Dataset: Bird Species Identification and Behavior Recognition in Videos,” Scientific Data, vol. 12, pp. 1-13, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[20] Pralhad Gavali, and J. Saira Banu, “A Novel Approach to Indian Bird Species Identification: Employing Visual-Acoustic Fusion Techniques for Improved Classification Accuracy,” Frontiers in Artificial Intelligence, vol. 8, pp. 1-15, 2025.
[CrossRef] [Google Scholar] [Publisher Link]
[21] Jie Hu, Li Shen, and Gang Sun, “Squeeze-and-Excitation Networks,” 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, pp. 7132-7141, 2018.
[CrossRef] [Google Scholar] [Publisher Link]
[22] Karen Simonyan, and Andrew Zisserman, “Very Deep Convolutional Networks for Large-Scale Image Recognition,” arXiv Preprint, pp. 1-14, 2014.
[CrossRef] [Google Scholar] [Publisher Link]
[23] Mingxing Tan, and Quoc Le, “Efficientnet: Rethinking Model Scaling for Convolutional Neural Networks,” International Conference on Machine Learning, pp. 6105-6114, 2019.
[Google Scholar] [Publisher Link]