IJET 2024 Vol.16(4): 217-221
DOI: 10.7763/IJET.2024.V16.1284
Plant Recognition on Highway Slopes Based on Improved Prototype Network Integrated Classification
Lailong Tang 1,
Xiangjie Zhai 1,2,
and
Zheng Kang 1,2
1.
School of Information Technology, Mongolian National University, Ulaanbaatar, Mongolian
2.
College of Art, Education Culture Law Institute of Mongolia, Ulaanbaatar, Mongolian
Email: zhejoys@foxmail.com (L.L.T.); zhaimingji@gmail.com (X.J.Z.); xyouzi8888@gmail.com (Z.K.)
*Corresponding author
Manuscript received September 24, 2024; revised October 14, 2024; accepted October 30, 2024; published November 19, 2024.
Abstract—Automatic identification of highway slope plants is of great significance for highway safety and ecological protection. However, the classification task of slope plants faces problems such as small sample size and large category diversity. In order to improve the accuracy and robustness of highway slope plant classification, in this paper, an attention-enhanced prototype network integrated classifier (Attention Enhancement Prototype Network Integrated Classifier, AEPEC) is proposed. First, the attention mechanism is introduced to enable the prototype network to focus on important regions in the image and extract more discriminative features. Secondly, the integrated learning method is used to integrate multiple attention enhancement prototype network models to further improve the model performance. This paper verifies the actual slope plant image dataset, and the experimental results show that the proposed method has achieved significant performance improvement in the slope plant classification task.
Keywords—slope plant classification, small sample learning, prototype network, attention mechanism and integrated learning
Cite: Lailong Tang, Xiangjie Zhai, and Zheng Kang, "Plant Recognition on Highway Slopes Based on Improved Prototype Network Integrated Classification,"
International Journal of Engineering and Technology, vol. 16, no. 4, pp. 217-221, 2024.
Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
