基于多层注意力机制的蜜蜂音频识别方法
廖越颖 （ 西南民族大学 计算机科学与人工智能学院 ）
黄闽英 （ 西南民族大学 计算机科学与人工智能学院 ）
https://doi.org/10.37155/2811-0617-0406-35Abstract
传统蜜蜂活动监测依赖人工检查，存在工作量大、易干扰蜜蜂等问题。计算机视觉和无线物联网技术的 发展推动了自动化监测，但图像识别受光照和环境影响，难以监测隐蔽天敌。相比之下，声音识别不受光线限制，能 捕捉生物声音，监测范围更广，精度更高。本研究提出两种方案提高天敌监测精度：一是采用声音识别技术，弥补图 像监测不足；二是引入多特征融合，结合SSD特征提取、散射图生成、FCN分析和HGCN注意力机制优化模型。实验 表明，该方法较传统方法效益提升5.8%至44%。
Keywords
蜜蜂天敌监测；声音识别；多特征融合Full Text
PDFReferences
[1]Soares B S, Luz J S, de Macêdo V F, et al. MFCCbased descriptor for bee queen presence detection[J]. Expert
Systems with Applications, 2022, 201: 117104.
[2]Phan T T H, Nguyen-Doan D, Nguyen-Huu D, et
al. Investigation on new Mel frequency cepstral coefficients
features and hyper-parameters tuning technique for bee sound
recognition[J]. Soft Computing, 2023, 27(9): 5873-5892.
[3]Borgianni L, Ahmed M S, Adami D, et al.
Spectrogram Based Bee Sound Analysis with DNNs: a
step toward Federated Learning approach[C]//2023 4th
International Symposium on the Internet of Sounds. IEEE,
2023: 1-8.
[4]Libal U, Biernacki P. MFCC selection by LASSO for
honey bee classification[J]. Applied Sciences, 2024, 14(2):
913.
[5]Kong Q, Cao Y, Iqbal T, et al. Panns: Largescale pretrained audio neural networks for audio pattern
recognition[J]. IEEE/ACM Transactions on Audio, Speech,
and Language Processing, 2020, 28: 2880-2894.
[6]Toffa O K, Mignotte M. Environmental sound
classification using local binary pattern and audio features
collaboration[J]. IEEE Transactions on Multimedia, 2020, 23:
3978-3985.
[7]Nanni L, Maguolo G, Brahnam S, et al. An ensembleof convolutional neural networks for audio classification[J].
Applied Sciences, 2021, 11(13): 5796.
[8]Verbitskiy S, Berikov V, Vyshegorodtsev V. Eranns:
Efficient residual audio neural networks for audio pattern
recognition[J]. Pattern Recognition Letters, 2022, 161: 38-44.
[9]Zgank A. Reduced MFCC Feature Extraction
Dimension for Acoustic Classification of Bee Swarm
Activity[C]//2022 ELEKTRO (ELEKTRO). IEEE, 2022: 1-4.
[10]Faiß M. InsectSet47 & InsectSet66: Expanded
datasets for automatic acoustic identification of insects
(Orthoptera and Cicadidae)[J]. Zenodo, 2023.
[11]Malaviya P, Kumar Y, Modi N. Advancements
in Environmental Sound Classification: Evaluating
Machine Learning and Deep Learning Approaches on the
UrbanSound8k[C]//2023 Seventh International Conference on
Image Information Processing (ICIIP). IEEE, 2023: 900-905.
[12]García-Ordás M T, Rubio-Martín S, BenítezAndrades J A, et al. Multispecies bird sound recognition using
a fully convolutional neural network[J]. Applied Intelligence,
2023, 53(20): 23287-23300.
[13]Gupta U, Gupta D. Least squares structural twin
bounded support vector machine on class scatter[J]. Applied
Intelligence, 2023, 53(12): 15321-15351.
[14]Yan Y. Using the Improved SSD Algorithm to
Motion Target Detection and Tracking[J]. Computational
Intelligence and Neuroscience, 2022, 2022(1): 1886964.
[15]Truong T H, Du Nguyen H, Mai T Q A, et al. A
deep learning-based approach for bee sound identification[J].
Ecological Informatics, 2023, 78: 102274.
Systems with Applications, 2022, 201: 117104.
[2]Phan T T H, Nguyen-Doan D, Nguyen-Huu D, et
al. Investigation on new Mel frequency cepstral coefficients
features and hyper-parameters tuning technique for bee sound
recognition[J]. Soft Computing, 2023, 27(9): 5873-5892.
[3]Borgianni L, Ahmed M S, Adami D, et al.
Spectrogram Based Bee Sound Analysis with DNNs: a
step toward Federated Learning approach[C]//2023 4th
International Symposium on the Internet of Sounds. IEEE,
2023: 1-8.
[4]Libal U, Biernacki P. MFCC selection by LASSO for
honey bee classification[J]. Applied Sciences, 2024, 14(2):
913.
[5]Kong Q, Cao Y, Iqbal T, et al. Panns: Largescale pretrained audio neural networks for audio pattern
recognition[J]. IEEE/ACM Transactions on Audio, Speech,
and Language Processing, 2020, 28: 2880-2894.
[6]Toffa O K, Mignotte M. Environmental sound
classification using local binary pattern and audio features
collaboration[J]. IEEE Transactions on Multimedia, 2020, 23:
3978-3985.
[7]Nanni L, Maguolo G, Brahnam S, et al. An ensembleof convolutional neural networks for audio classification[J].
Applied Sciences, 2021, 11(13): 5796.
[8]Verbitskiy S, Berikov V, Vyshegorodtsev V. Eranns:
Efficient residual audio neural networks for audio pattern
recognition[J]. Pattern Recognition Letters, 2022, 161: 38-44.
[9]Zgank A. Reduced MFCC Feature Extraction
Dimension for Acoustic Classification of Bee Swarm
Activity[C]//2022 ELEKTRO (ELEKTRO). IEEE, 2022: 1-4.
[10]Faiß M. InsectSet47 & InsectSet66: Expanded
datasets for automatic acoustic identification of insects
(Orthoptera and Cicadidae)[J]. Zenodo, 2023.
[11]Malaviya P, Kumar Y, Modi N. Advancements
in Environmental Sound Classification: Evaluating
Machine Learning and Deep Learning Approaches on the
UrbanSound8k[C]//2023 Seventh International Conference on
Image Information Processing (ICIIP). IEEE, 2023: 900-905.
[12]García-Ordás M T, Rubio-Martín S, BenítezAndrades J A, et al. Multispecies bird sound recognition using
a fully convolutional neural network[J]. Applied Intelligence,
2023, 53(20): 23287-23300.
[13]Gupta U, Gupta D. Least squares structural twin
bounded support vector machine on class scatter[J]. Applied
Intelligence, 2023, 53(12): 15321-15351.
[14]Yan Y. Using the Improved SSD Algorithm to
Motion Target Detection and Tracking[J]. Computational
Intelligence and Neuroscience, 2022, 2022(1): 1886964.
[15]Truong T H, Du Nguyen H, Mai T Q A, et al. A
deep learning-based approach for bee sound identification[J].
Ecological Informatics, 2023, 78: 102274.
Copyright © 2025 廖越颖,黄闽英
Publishing time:2025-06-30
This work is licensed under a Creative Commons Attribution 4.0 International License
Email
Location