Audio Classification of Water Supply Pipeline Leakage Based on MobileNetV3 Convolutional Neural Network

CHEN Shuangye; XU Leiheng; HUANG Chengyi; ZHANG Zhiwu; ZHANG Lin; HAN Mo

doi:10.11936/bjutxb2022090017

Journal of Beijing University of Technology > 2024 > 50(7): 797-804. > DOI: 10.11936/bjutxb2022090017

CHEN Shuangye, XU Leiheng, HUANG Chengyi, ZHANG Zhiwu, ZHANG Lin, HAN Mo. Audio Classification of Water Supply Pipeline Leakage Based on MobileNetV3 Convolutional Neural Network[J]. Journal of Beijing University of Technology, 2024, 50(7): 797-804. DOI: 10.11936/bjutxb2022090017

Citation:

PDF (2014 KB)

Audio Classification of Water Supply Pipeline Leakage Based on MobileNetV3 Convolutional Neural Network

1.
Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China
2.
Beijing Water Supply Group Yutong Municipal Engineering Co., Ltd., Beijing 100142, China

More Information

Received Date: September 14, 2022
Revised Date: January 11, 2023
Available Online: May 27, 2024

Graphical Abstract

Abstract

Abstract

To accurately identify the leakage sound of the urban water supply network, this paper proposes an audio classification and recognition method for water supply pipeline leakage based on MobileNetV3. First, the audio files in the ROPP dataset were enhanced offline, the leakage signal was converted into a logarithmic Mel spectrogram and the spectral subtraction was used to achieve data noise reduction. The Attention Mechanism module and MobileNetV3 neural network training were used to extract image features. Finally, the Softmax function was used to classify the leaked audio. The experimental results show that this method can make the classification precision of water leakage category reach 99.40%, and the recall rate reaches 99.20%.
- sound event classification,
- water pipe leak detection,
- MobileNetV3,
- data enhancement,
- spectral subtraction,
- squeeze and excitation

FullText(HTML)

References (24)

References

[1]	贾亚杰. 供水管道漏点检测技术优化开发研究[D]. 太原: 太原理工大学, 2016: 1. JIA Y J. Optimization research and development pipeline leak location technology of water supply project[D]. Taiyuan: Taiyuan University of Technology, 2016: 1. (in Chinese)
[2]	吕杰. 基于DSP的数字式漏水探测仪[D]. 上海: 上海交通大学, 2008: 1-4. LÜ J. Digital instrument of detecting water leakage based on DSP[D]. Shanghai: Shanghai Jiao Tong University, 2008: 1-4. (in Chinese)
[3]	陈天娇, 曾娟, 谢成军, 等. 基于深度学习的病虫害智能化识别系统[J]. 中国植保导刊, 2019, 39(4): 26-34. https://www.cnki.com.cn/Article/CJFDTOTAL-ZBJS201904006.htm CHEN T J, ZENG J, XIE C J, et al. Intelligent identification system of disease and insect pests based on deep learning[J]. China Plant Protection Guide, 2019, 39(4): 26-34. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZBJS201904006.htm
[4]	大连大学. 一种轻小型网络自适应番茄病害特征提取方法: 113344009A[P/OL]. 2021-09-03[2023-01-12]. https://cprs.patentstar.com.cn/Search/Detail?ANE=9IGH9CHA5ADA4CBA2BAA4EAAADIA9GHG9HDC9BFA9CGF6EBA
[5]	XIE Y C, XIAO Y C, LIU X Y, et al. Time-frequency distribution map-based convolutional neural network (CNN) model for underwater pipeline leakage detection using acoustic signals[J/OL]. Sensors, 2020, 20: 5040[2023-01-12]. https://www.semanticscholar.org/paper/Time-Frequency-Distribution-Map-Based-Convolutional-Xie-Xiao/ec23131a86b3534983c8da8031e0483496a860a8. DOI: 10.3390/s20185040.
[6]	SHUKLA H, PIRATLA K. Leakage detection in water pipelines using supervised classification of acceleration signals[J/OL]. Automation in Construction, 2020, 117(4): 103256[2023-01-12]. https://linkinghub.elsevier.com/retrieve/pii/S0926580519310301.
[7]	KANG J, PARK Y J, LEE J, et al. Novel leakage detection by ensemble CNN-SVM and graph-based localization in water distribution systems[J/OL]. IEEE Transactions on Industrial Electronics, 2018, 65(5): 4279-4289[2023-01-12]. https://ieeexplore.ieee.org/document/8074786.
[8]	CHEN S Y, HU X, XU L H, et al. Construction and experimental research on leakage sound dataset of urban water supply pipeline[C/OL]. 2021 China Automation Congress. Piscataway, NJ: IEEE, 2021: 5385-5390[2023-01-12]. https://ieeexplore.ieee.org/document/9727424.
[9]	吴乔域. 基于跃层注意力机制的轻量化卷积神经网络的情绪识别方法和系统研究[D/OL]. 杭州: 浙江大学, 2021: 53-56[2023-01-12]. https://kns.cnki.net/kcms/detail/detail.aspx?DbCode=CMFD&Filename=1021113360.nh. WU Q Y. Research on technologies and system of emotion recognition based on lightweight skip-layer attention convolution neural network[D/OL]. Hangzhou: Zhejiang University, 2021: 53-56[2023-01-12]. https://kns.cnki.net/kcms2/article/abstract?v=3uoqIhG8C44Mq-Kzu8QOorG3AZnu--iM-vMcHu5qQTHWR3fC8LTAzSHatU qMqnV72TlxrHsQi8_sdtBTe_XlLAiqOE4fbWNQ&uniplatform=NZKPT&language=en. (in Chinese)
[10]	MESAROS A, HEITTOLA T, VIRTANEN T. A multi-device dataset for urban acoustic scene classification[J/OL]. arXiv: 1807.09840v2[2023-01-12]. https://arxiv.org/abs/1807.09840v2.
[11]	MCFEE B, RAFFEL C, LIANG D W, et al. Librosa: audio and music signal analysis in python[C/OL]//Proceedings of the 14th Python in Science Conference(SciPy2015)[2023-01-12]. http://conference.scipy.org.s3-website-us-east-1.amazonaws.com/proceedings/scipy2015/brian_mcfee.html. DOI: 10.25080/Majora-7b98e3ed-003.
[12]	淄博艾吉电气有限公司. ETZ-1A(C)电子听诊器[Z/OL]. [2023-01-12]. http://www.exagiga.com/productinfo/583423.html.
[13]	BOLL S. Suppression of acoustic noise in speech using spectral subtraction[J/OL]. IEEE Transactions on Acoustics, Speech, and Signal Processing, 1979, 27(2): 113-120[2023-01-12]. https://ieeexplore.ieee.org/document/1163209.
[14]	郑成诗, 胡笑浒, 周翊, 等. 基于噪声谱结构特性的谱减法[J/OL]. 声学学报, 2010, 35(2): 215-222[2023-01-12]. https://kns.cnki.net/KCMS/detail/detail.aspx?filename=XIBA201002019&dbname=CJFD&dbcode=CJFQ. ZHENG C S, HU X G, ZHOU Y, et al. Spectral subtraction based on the structure of noise power spectral density[J/OL]. Acta Acustica, 2010, 35(2): 215-222[2023-01-12]. http://en.cnki.com.cn/Article_en/CJFDTOTAL-xiba201002019.htm. (in Chinese)
[15]	王怡婷. 基于深度神经网络的语音增强模型研究[D]. 上海: 东华理工大学, 2021: 1-2. WANG Y T. Research on speech enhancement model based on deep neural network. [D]. Shanghai: East China University of Technology, 2021: 1-2. (in Chinese)
[16]	BEROUTI M, SCHWARTZ R, MAKHOUL J. Enhancement of speech corrupted by acoustic noise[C/OL]// IEEE International Conference on Acoustics, Speech, and Signal Processing. Piscataway, NJ: IEEE, 1979: 208-211[2023-01-12]. https://ieeexplore.ieee.org/document/1170788.
[17]	何超. 会议电话中的实时语音降噪[D]. 南京: 东南大学, 2019: 9-11. HE C. Real-time noise reduction for conference telephone[D]. Nanjing: Southeast University, 2019: 9-11. (in Chinese)
[18]	陈源, 蔡文郁, 胡毕炜. 微型水声传感网水下传感器节点设计[J]. 电子科技, 2019, 32(7): 71-75. https://www.cnki.com.cn/Article/CJFDTOTAL-DZKK201907015.htm CHEN Y, CAI W Y, HU B W. Underwater sensor node design of micro underwater acoustic sensor network[J]. Electronic Science and Technology, 2019, 32(7): 71-75. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-DZKK201907015.htm
[19]	段嘉伟, 杨迪, 许佳豪, 等. 基于MobileNetV3与注意力机制的脸型分类研究[J]. 长江信息通信, 2022, 35(1): 13-16. https://www.cnki.com.cn/Article/CJFDTOTAL-HBYD202201005.htm DUAN J W, YANG D, XU J H, et al. Face shape classification based on MobileNetV3 and attention mechanism[J]. Changjiang Information & Communications, 2022, 35(1): 13-16. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-HBYD202201005.htm
[20]	HU J, SHEN L, SUN G, et al. Squeeze-and-excitation networks[C/OL]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, NJ: IEEE, 2018: 7132-7141[2023-01-12]. https://ieeexplore.ieee.org/document/8578843. DOI: 10.1109/CVPR.2018.00745.
[21]	王璇, 康硕, 朱伟东. 基于改进CenterNet的AFP铺层表面缺陷检测[J/OL]. 红外与激光工程, 2021, 50(10): 20210011[2023-01-12]. https://kns.cnki.net/KCMS/detail/detail.aspx?filename=HWYJ202110025&dbname=CJFD&dbcode=CJFQ. WANG X, KANG S, ZHU W D. Defect detection of laminated surface in the automated fiber placement process based on improved CenterNet[J/OL]. Infrared and Laser Engineering, 2021, 50(10): 20210011[2023-01-12]. http://en.cnki.com.cn/Article_en/CJFDTOTAL-hwyj202110025.htm. (in Chinese)
[22]	刘洋, 余甜, 丁艺. 一种新的基于最大概率路径的中文分词[J]. 计算机与数字工程, 2022, 50(3): 591-596. https://www.cnki.com.cn/Article/CJFDTOTAL-JSSG202203026.htm LIU Y, YU T, DING Y. A new Chinese word segmentation based on maximum probability path[J]. Computer & Digital Engineering, 2022, 50(3): 591-596. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-JSSG202203026.htm
[23]	闫壮壮, 闫学慧, 石嘉, 等. 基于深度学习的大豆豆荚类别识别研究[J]. 作物学报, 2020, 46(11): 1771-1779. https://www.cnki.com.cn/Article/CJFDTOTAL-XBZW202011014.htm YAN Z Z, YAN X H, SHI J, et al. Classification of soybean pods using deep learning[J]. Acta Agronomica Sinica, 2020, 46(11): 1771-1779. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-XBZW202011014.htm
[24]	张珂, 王新胜, 郭玉荣, 等. 人脸年龄估计的深度学习方法综述[J]. 中国图象图形学报, 2019, 24(8): 1215-1230. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201908002.htm ZHANG K, WANG X S, GUO Y R, et al. Survey of deep learning methods for face age estimation[J]. Journal of Image and Graphics, 2019, 24(8): 1215-1230. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGTB201908002.htm

[1]	LING Xiaoming, CHEN Hongyan, ZHANG Xiaoyu, ZHANG Zhen. Speaker Recognition Algorithm Based on ASP-SERes2Net[J]. Journal of Beijing University of Technology, 2025, 51(1): 42-50. DOI: 10.11936/bjutxb2023060027
[2]	LIU Hongye, JIN Jiaoyan, FAN Yanping, MA Sha, LIU Zenghua, HE Cunfu. Research on Lamb Wave Imaging of a Sessile Water Droplet With Chirp Excitation[J]. Journal of Beijing University of Technology, 2018, 44(5): 749-756. DOI: 10.11936/bjutxb2017060001
[3]	YANG Xiaojing, YANG Yang, LI Huaizhou, ZHONG Ning. Different Strategies in Solving Arithmetic Calculation Problems[J]. Journal of Beijing University of Technology, 2017, 43(4): 587-593. DOI: 10.11936/bjutxb2016050067
[4]	JI Junzhong, GAO Guangxuan. Detecting Functional Module Method Based on Cultural Algorithm in Protein-protein Interaction Networks[J]. Journal of Beijing University of Technology, 2017, 43(1): 13-21. DOI: 10.11936/bjutxb2016100034
[5]	JIANG Xiao-wei, LI Zhi-ming, LI Yun, YU Xin, LIU Chu-sheng. Single Axis Variable Trajectory Screen Transport Mechanism and Response Spectrum Analysis of Exciting Point[J]. Journal of Beijing University of Technology, 2013, 39(11): 1660-1666.
[6]	WANG Shu-wei, CHEN Yan-yan, BO Ying-chun. Pressure Model of Contact Interface Between Vehicle Tires and Highway Pavement Based on Modular Neural Network[J]. Journal of Beijing University of Technology, 2013, 39(2): 245-250.
[7]	ZHUO Li, XIAO Zhu, WANG Su-yu, LI Xiao-guang, ZHU Qing. A Spectral Imagery Compression Algorithm Based on Spectral Prediction and Wavelet Transform[J]. Journal of Beijing University of Technology, 2010, 36(8): 1037-1043.
[8]	PENG Xi-rong, LU Xin-ying, SUI Yun-kang, CHEN Zhao-yuan. A Damage Index Based on Strain Measurement for Structures Under Environmental Excitation[J]. Journal of Beijing University of Technology, 2008, 34(5): 449-453.
[9]	LU Zhi-yuan, LI De-sheng, WANG Wei, LIU Zhe, GUO Qiao-hong. Suspending Globe With Hybrid-excited Magnets Based on PDF[J]. Journal of Beijing University of Technology, 2007, 33(10): 1021-1025.
[10]	ZHANG Yanqing, ZHOU Mingfang. Study on the Forming Process of Squeezed Pile[J]. Journal of Beijing University of Technology, 2001, 27(1): 54-56.

Cited By

Cited by

Periodical cited type(1)

唐军，张政. 城市自来水厂供水管道铺设与安装施工技术研究. 科学技术创新. 2025(04): 127-130 .

Other cited types(0)

Get Citation

PDF

XML

Article views PDF downloads Cited by(1)

Turn off MathJax

Article Contents

Abstract

References

Audio Classification of Water Supply Pipeline Leakage Based on MobileNetV3 Convolutional Neural Network