Advancing Winged Animal Classification Through Image Analysis

Prasetya Triputra Nugraha, Imam Yuadi

Abstract


This study is to assess how well two classification algorithms, Support Vector Machine (SVM) and Logistic Regression, work with deep learning-based feature extraction techniques, including Inception V3, VGG-16, and VGG-19. The methodology comprised preprocessing a collection of photos of flying animals, using the three convolutional neural network (CNN) designs to extract features, and applying the two algorithms to do classification. AUC, Classification Accuracy (CA), F1 Score, Precision, Recall, and MCC were among the important metrics used to assess the models. According to the findings, Inception V3 performed better than VGG-16 and VGG-19 on every parameter, with Logistic Regression obtaining nearly flawless scores (AUC = 1.000, CA = 0.987, F1 = 0.987). Although it was marginally less effective than Logistic Regression (AUC = 0.998, CA = 0.943, F1 = 0.946), SVM also did well with Inception V3. The feature extraction techniques that performed the worst were VGG-16 and SVM in particular (CA = 0.890, F1 = 0.891). These results highlight the effectiveness of Logistic Regression for classification in this setting and the improved multi-scale feature extraction capabilities of Inception V3. This study demonstrates how effective classifiers and cutting-edge CNN architectures, such as Inception V3, may be combined to automatically classify winged animals.

Keywords


Animal; Classification; Convilutional Neural Network; Logistic Regression; Support Vector Machine.

Full Text:

PDF

References


Ahmed, M. F., & Khan, A. M. (2021). Identification of plant species by deep learning and providing as a mobile application. Sakarya University Journal of Computer and Information Sciences, 3(3), 273–282.

Ahmed, T., & Khan, R. (2021). Efficient bird species classification using transfer learning and fine-tuned CNNs. Ecological Informatics, 62, 101273.

Brown, C., Smith, J., & Lee, D. (2019). Deep learning-based ecological analysis of camera trap images is impacted by training data quality and size. Ecological Informatics, 53, 100976.

Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009). ImageNet: A large-scale hierarchical image database. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 248–255).

Garcia, M. A., & Wang, S. (2022). A new deep learning system for wild plants classification and species identification using leaves and fruits. Journal of Plant Sciences, 10(2), 45–56.

Han, J., Kamber, M., & Pei, J. (2011). Data mining: Concepts and techniques (3rd ed.). Morgan Kaufmann.

Hosmer, D. W., Lemeshow, S., & Sturdivant, R. X. (2013). Applied logistic regression (3rd ed.). John Wiley & Sons.

Hsu, C. W., Chang, C. C., & Lin, C. J. (2016). A practical guide to support vector classification. Department of Computer Science, National Taiwan University.

Jones, R., & Lee, S. (2021). Machine learning and deep learning—A review for ecologists. Methods in Ecology and Evolution, 12(10), 2006–2021.

Jones, R., & Patel, M. (2021). Applications for deep learning in ecology. Methods in Ecology and Evolution, 12(11), 2153–2164.

Li, X., Zhang, C., & Liu, Y. (2020). A comprehensive survey of animal identification: Exploring data, methodologies, and applications. International Journal of Computer Vision, 128(6), 1690–1717.

Li, Y., Fang, L., Mei, Z., & Wang, J. (2020). Deep learning for ecological image classification: A case study on bird species identification. Ecological Informatics, 56, 101056.

Patel, M., Wang, Y., & Garcia, M. A. (2020). Deep learning enables satellite-based monitoring of large mammals across fine spatial scales. Nature Communications, 11, 5435.

Simonyan, K., & Zisserman, A. (2015). Very deep convolutional networks for large-scale image recognition. In Proceedings of the International Conference on Learning Representations (ICLR).

Smith, L. N., & Topin, N. (2020). Super-convergence: Very fast training of neural networks using large learning rates. In Proceedings of Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications (Vol. 11006, p. 1100612).

Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016). Rethinking the Inception architecture for computer vision. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 2818–2826).

Wang, Y., Garcia, M. A., & Patel, M. (2022). Deep learning-based ecological analysis of camera trap images is impacted by training data quality and size. Ecological Informatics, 68, 101505.

Wang, Y., Zhao, L., & Liu, Q. (2022). An ensemble learning framework for robust bird species classification in complex environments. Pattern Recognition Letters, 157, 28–36.

Zhang, C., Li, X., & Liu, Y. (2019). Bird species classification from an image using VGG-16 network. International Journal of Computer Vision, 127(6–7), 671–682.

Zhang, Y., Zhou, W., & Zhang, X. (2019). Bird species recognition in natural scenes using convolutional neural networks. Computers and Electronics in Agriculture, 157, 447–455.




DOI: https://doi.org/10.51849/j-p3k.v7i2.994

Refbacks

  • There are currently no refbacks.