Clasificación del sustrato del fondo de nubes de puntos LiDAR aerotransportado considerando la selección óptima de características

  • role: First author第一作者
  • Affiliation:

    College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266590, China

    Key Laboratory of Ocean Geomatics, Ministry of Natural Resources of China, Qingdao 266590, China

    Key Laboratory of Submarine Geosciences, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou 310012, China

  • Email:sudianpeng@126.com
  • Introduction:宿殿鹏,研究方向为机载LiDAR测深数据处理与应用。E-mail: sudianpeng@126.com
SU Dianpeng123,  
  • Affiliation:

    College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266590, China

HUANG Yu1,  
  • role: Corresponding author通信作者
  • Affiliation:

    College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266590, China

    Key Laboratory of Ocean Geomatics, Ministry of Natural Resources of China, Qingdao 266590, China

  • Email:flyang@126.com
  • Introduction:阳凡林,研究方向为海底地形测量和海洋定位导航。 E-mail: flyang@126.com
YANG Fanlin12*,  
  • Affiliation:

    Key Laboratory of Ocean Geomatics, Ministry of Natural Resources of China, Qingdao 266590, China

    Key Laboratory of Submarine Geosciences, Second Institute of Oceanography, Ministry of Natural Resources, Hangzhou 310012, China

ZHAO Dineng23,  
  • Affiliation:

    College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266590, China

    Key Laboratory of Ocean Geomatics, Ministry of Natural Resources of China, Qingdao 266590, China

YANG Anxiu12,  
  • Affiliation:

    College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266590, China

LIU Jiaoyang1

resumen

La clasificación del sustrato marino basada en la tecnología ALB (Batimetría LiDAR Aerotransportada) puede proporcionar datos básicos para el desarrollo de recursos marinos en aguas poco profundas, la protección del medio ambiente marino y la construcción de ingeniería marina, y es de gran importancia para las actividades marítimas y la investigación científica marina. Para abordar el problema de la redundancia de características en la clasificación del sustrato marino ALB, este artículo propone un algoritmo de clasificación del sustrato que considera la selección óptima de características de forma de onda y topográficas. Basado en la extracción de características de forma de onda y topográficas, se construyó un modelo de selección de características Relief-F que, al calcular la contribución de cada característica en la clasificación del sustrato, realiza una selección múltiple de características; luego, utilizando tres clasificadores supervisados: Bosque Aleatorio RF (Random Forest), Máquina de Vectores de Soporte SVM (Support Vector Machine) y Red Neuronal BPNN (Back Propagation Neural Network), se realiza la clasificación para extraer cinco tipos de sustrato: arrecifes de coral, grava, arena, vegetación y zona costera. Para verificar la efectividad del método de clasificación propuesto, se realizaron experimentos utilizando datos ALB medidos en la isla Ganquan del archipiélago de Xisha, los resultados muestran que después de la selección de características utilizando el algoritmo Relief-F, la precisión de clasificación de RF, SVM y BPNN aumentó en un 1,1%, 1,1% y 2,7% respectivamente; entre ellos, la clasificación con Bosque Aleatorio presenta una mayor precisión, con una precisión general OA (Overall Accuracy) y coeficiente Kappa de 95,36% y 0,94 respectivamente. Los resultados de este estudio pueden proporcionar un soporte técnico efectivo para las necesidades de clasificación del sustrato marino en campos como la ingeniería marina.

palabra clave

LiDAR aerotransportado; clasificación del sustrato; características de forma de onda; características topográficas; modelo de selección de características Relief-F; procesamiento de imágenes; océano

References

  1. 1.
    Abdallah H, Baghdadi N, Bailly J S, Pastol Y and Fabre F. 2012. Wa-LiD: a new LiDAR simulator for waters. IEEE Geoscience and Remote Sensing Letters, 9(4): 744-748
  2. 2.
    Alexander C, Tansey K, Kaduk J, Holland D and Tate N J. 2010. Backscatter coefficient as an attribute for the classification of full-waveform airborne laser scanning data in urban areas. ISPRS Journal of Photogrammetry and Remote Sensing, 65(5): 423-432
  3. 3.
    Ayustina R, Aulia Z, Mustakin H, Alam F, Amron A, Yuwono D, Ahmad T, Prayogo A and Sari F. 2018. Classification of shallow water seabed profile based on Landsat 8 imagery and in-situ data. Case study in Gili Matra Island Lombok, Indonesia. E3S Web of Conferences, 47: 04002
  4. 4.
    Chen J, Mao X C, Liu Z K and Deng H. 2020. Three-dimensional metallogenic prediction based on random forest classification algorithm for the Dayingezhuang gold deposit. Geotectonica et Metallogenia, 44(2): 231-241
  5. 5.
    Chen J B, Wu Z Y, Zhao D N, Zhou J Q, Li S J and Shang J H. 2017. Back propagation neural network classification of sediment seabed acoustic sonar images based on particle swarm optimization algorithms. Acta Oceanologica Sinica, 39(9): 51-57
  6. 6.
    Chen K. 2021. Research on Feature Selection Methods for High-Dimensional Classification. Ji’nan: Shandong University
  7. 7.
    Dai M F, Xing S, Xu Q, Li P C and Chen K. 2022. Semantic segmentation of airborne LiDAR point cloud based on multi-feature fusion and geometric convolution. Journal of Image and Graphics, 27(2): 574-585
  8. 8.
    Eren F, Pe'Eri S, Rzhanov Y and Ward L. 2018. Bottom characterization by using airborne lidar bathymetry (ALB) waveform features obtained from bottom return residual analysis. Remote Sensing of Environment, 206: 260-274
  9. 9.
    Ge Q, Ruan F X, Qiao B J, Zhang Q, Zuo X Y and Dang L X. 2021. Side-scan sonar image classification based on style transfer and pre-trained convolutional neural networks. Electronics, 10(15): 1823
  10. 10.
    Hou D J, Wu C, Li C Z, Guo S M and Wu J J. 2019. Energy calibration and performance testing of LEGe detector. The Journal of Engineering, 2019(23): 9031-9034
  11. 11.
    Huang L L, Tang J, Sun D D and Luo B. 2012. Feature selection algorithm based on multi-label ReliefF. Journal of Computer Applications, 32(10): 2888-2890, 2898
  12. 12.
    Joanes D N and Gill C A. 1998. Comparing measures of sample skewness and kurtosis. Journal of the Royal Statistical Society: Series D (The Statistician), 47(1): 183-189
  13. 13.
    Lai X D and Zheng M. 2015. A denoising method for LiDAR full-waveform data. Mathematical Problems in Engineering, 2015: 164318
  14. 14.
    Lee A L, Bartle K D and Lewis A C. 2001. A model of peak amplitude enhancement in orthogonal two-dimensional gas chromatography. Analytical Chemistry, 73(6): 1330-1335
  15. 15.
    Li L F, Liu F, Wang C and Chen L S. 2012. Measurement and control of residual amplitude modulation in optical phase modulation. Review of Scientific Instruments, 83(4): 043111
  16. 16.
    Liu Y M, Deng R R, Qin Y and Liang Y H. 2017. Data processing methods and applications of airborne LiDAR bathymetry. Journal of Remote Sensing, 21(6): 982-995
  17. 17.
    Liu Y X, Guo K, He X F, Xu W X and Feng Y K. 2017. Research progress of airborne laser bathymetry technology. Geomatics and Information Science of Wuhan University, 42(9): 1185-1194
  18. 18.
    Matsatsinis N F and Samaras A P. 2000. Brand choice model selection based on consumers’ multicriteria preferences and experts’ knowledge. Computers and Operations Research, 27(7/8): 689-707
  19. 19.
    Qi C, Su D P, Wang X K, Wang M W, Shi B and Yang F L. 2019. Fitting algorithm for airborne laser bathymetric waveforms based on layered heterogeneous model. Infrared and Laser Engineering, 48(2): 206004
  20. 20.
    Qi C, Zhou F N, Wu J W, Su D P, Wang X K and Yang F L. 2021. Extraction method for diffuse attenuation coefficient based on airborne LiDAR bathymetric water column waveform. Acta Oceanologica Sinica, 43(1): 147-154
  21. 21.
    Qiao J G, Liu X P and Zhang Y H. 2011. Land cover classification using LiDAR height texture and ANNs. Journal of Remote Sensing, 15(3): 539-553
  22. 22.
    Su D P, Yang F L, Ma Y, Zhang K, Huang J and Wang M W. 2019. Classification of coral reefs in the South China Sea by combining airborne LiDAR bathymetry bottom waveforms and bathymetric features. IEEE Transactions on Geoscience and Remote Sensing, 57(2): 815-828
  23. 23.
    Tang Q H, Li J, Zhou X H, Lu K and Zhang Z Z. 2014. Seabed sonar image analysis and acoustic seabed classification in the south of the Cheju Island. Acta Oceanologica Sinica, 36(7): 133-141
  24. 24.
    Velasco J, Molina I, Martinez E, Arquero A and Prieto J F. 2014. Sea bottom classification by means of bathymetric LIDAR data. IEEE Latin America Transactions, 12(4): 590-595
  25. 25.
    Xu C. 2014. Study on Seabed Classification Technology Based on Multibeam Bathymetry Sonar. Harbin: Harbin Engineering University
  26. 26.
    Yang F L, Zhu Z R, Li J B, Feng C K, Xing Z and Wu Z Y. 2021. Seafloor classification based on combined multibeam bathymetry and backscatter using deep convolution neural network. Acta Geodaetica et Cartographica Sinica, 50(1): 71-84
  27. 27.
    Yang M L, Fan Y G and Li B Y. 2020. Research on dimensionality reduction and classification of hyperspectral images based on LDA and ELM. Journal of Electronic Measurement and Instrumentation, 34(5): 190-196
  28. 28.
    Zang Z, Lin H and Yang M H. 2014. Dimensional reduction and classification of hyperspectral data for tree species using PCA algorithm. Science of Surveying and Mapping, 39(2): 146-149
  29. 29.
    Zhang H. 2012. Characteristic weighting FCM based on improved ReliefF algorithm. Shipboard Electronic Countermeasure, 35(1): 79-82, 85
  30. 30.
    Zhao Y X and Zhao T. 2020. Survey of the intelligent seabed sediment classification technology based on sonar images. CAAI Transactions on Intelligent Systems, 15(3): 587-600

Leer el texto completo

The above content is generated by Large Model Translation. The translated content is for reference only. We do not assume any commercial or legal responsibilty for any consequences arising from the use of our website