# An Accurate Recognition Method for Landslides Based on a Semi-Supervised Generative Adversarial Network: A Case Study in Lanzhou City

^{*}

## Abstract

**:**

^{2}, with an accuracy rate of 83%. Therefore, the generated results are accurate and reliable, and show that SSGAN can better distinguish landslides from non-landslides in an image, under the condition of obtaining a large number of unmarked environmental features; enhance the effect of landslide classification in complex geographical environment; and then put forward effective suggestions for the prevention and control of landslides and geological disasters in the main urban area of Lanzhou.

## 1. Introduction

^{2}, as the study area. The area is located in the transition zone from the Qinghai–Tibet Plateau to the Loess Plateau, with complex terrain and geological environment, which is a typical disaster-prone area [17]. At the same time, Lanzhou has a dense transportation network and large population flows. Frequent human activities such as engineering construction and mineral resource exploitation have also become important causes of landslide disasters [18,19].

## 2. Study the Regional Profile and Data Sources

#### 2.1. Study Area Description

#### 2.2. Data Sources

#### 2.2.1. Sentinel-2A Data

#### 2.2.2. Landslide Influencing Factors Data

## 3. Methods for Landslide Identification

#### 3.1. Semi-Supervised Learning

#### 3.2. Generative Adversarial Network

#### 3.3. Semi-Supervised Generative Adversarial Network Model Construction

#### 3.4. Verification Method

#### 3.4.1. Landslide Factor Analysis

^{2}represents the complex correlation coefficient of a specific independent variable when a regression analysis is conducted with the other independent variables. When this variable exhibits a high degree of correlation with the remaining variables, the value of A

^{2}approaches 1, resulting in a larger value for VIF (variance inflation factor). If the VIF exceeds 10 and its reciprocal, tolerance (T), is less than 0.1, it indicates the presence of multicollinearity.

#### 3.4.2. Precision Analysis

## 4. Accurate Recognition of Landslides Based on Semi-Supervised Generative Adversarial Network

#### 4.1. Dataset Building

#### 4.2. Understanding and Analysis of Semi-Supervised Generative Adversarial Processes

## 5. Results and Discussion

#### 5.1. Characteristic Analysis of Landslide Factors

#### 5.2. Landslide Recognition Results

^{2}. SSGAN identified 160 landslides in the study area, the minimum landslide area identified is 68 m

^{2}and the maximum is 0.68 km

^{2}, the total area is 10.328 km

^{2}, which is 3.139 km

^{2}more than the recognition results of the landslide constraints, as shown in Figure 8b. From Figure 8, it is evident that more small landslides were identified without the adversarial learning model. The SSGAN recognition results align closely with the distribution of the sample landslides, providing more detailed information about the landslide edges, and even capturing some individual landslides that exceed the original landslide areas. Both methods identified more landslides than the original sample.

## 6. Conclusions

^{2}were identified in the research area, which to some extent improves the landslide identification accuracy. In summary, the semi-supervised generative adversarial network method is a highly reliable landslide identification approach, which can provide a reference for landslide hazard identification in the main urban area of Lanzhou and similar working conditions.

## Author Contributions

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Conflicts of Interest

## References

- Yue, S. Research on Landslide Identification Technology Based on Multi-Source Remote Sensing Data. Master’s Thesis, Chang’an University, Chang’an, China, 2023. [Google Scholar]
- Wang, Y.; Fang, Z.; Niu, R.; Peng, L. Analysis of landslide susceptibility based on deep learning. J. Geo-Inf. Sci.
**2021**, 23, 2244–2260. [Google Scholar] - Niu, P. Evaluation of Landslide Susceptibility in Zhouqu County Based on a Comprehensive Index Model. Master’s Thesis, Hebei University of Geosciences, Shijiazhuang, China, 2021. [Google Scholar]
- Chen, S.; Miao, Z.; Wu, L. Simplified Newmark displacement method for rapid assessment of seismic landslide hazard based on modified rock and soil strength parameters. J. Seismol.
**2022**, 44, 512–527. [Google Scholar] - Huang, F.; Huang, J.; Jiang, S.; Zhou, C. Landslide displacement prediction based on multivariate chaotic model and extreme learning machine. Eng. Geol.
**2017**, 218, 173–186. [Google Scholar] [CrossRef] - Ma, Z.; Mei, G.; Piccialli, F. Machine learning for landslides prevention: A survey. Neural Comput. Appl.
**2021**, 33, 10881–10907. [Google Scholar] [CrossRef] - Dou, J.; Yunus, A.P.; Tien Bui, D.; Merghadi, A.; Sahana, M.; Zhu, Z.; Chen, C.; Khosravi, K.; Yang, Y.; Pham, B. Assessment of advanced random forest and decision tree algorithms for modeling rainfall-induced landslide susceptibility in the Izu-Oshima Volcanic Island, Japan. Sci. Total Environ.
**2019**, 662, 332–346. [Google Scholar] [CrossRef] [PubMed] - Wu, X.; Shen, S.; Niu, R. Applying PSO-SVM model to predict landslide susceptibility under GIS support. J. Wuhan Univ. (Inf. Sci. Ed.)
**2016**, 41, 665–671. [Google Scholar] - Zhang, T.; Han, L.; Han, J.; Li, X.; Zhang, H.; Wang, H. Assessment of Landslide Susceptibility Using Integrated Ensemble Fractal Dimension with Kernel Logistic Regression Model. Entropy
**2019**, 21, 218. [Google Scholar] [CrossRef] [PubMed] - He, Y.; Zhao, Z.; Yang, W.; Yan, H.; Wang, W.; Sheng, Y.; Zhang, L.; Tao, L. A unified network of information considering superimposed landslide factors sequence and pixel spatial neighbourhood for landslide susceptibility mapping. Int. J. Appl. Earth Obs. Geoinf.
**2021**, 104, 102508. [Google Scholar] [CrossRef] - Ju, Y. Early Recognition of Loess Landslide Based on UAV Photogrammetry A Case Study of Heifangtai. Master’s Thesis, Chengdu University of Technology, Chengdu, China, 2017. [Google Scholar]
- Ghorbanzadeh, O.; Blaschke, T.; Gholamnia, K.; Meena, S.; Tiede, D.; Aryal, J. Evaluation of different machine learning methods and deep-learning convolutional neural networks for landslide detection. Remote Sens.
**2019**, 11, 196. [Google Scholar] [CrossRef] - Fang, R.; Liu, Y.; Huang, Z. Summary of regional landslide risk assessment methods based on machine learning. Chin. J. Geol. Hazard Prev.
**2021**, 4, 5–12. [Google Scholar] - Qiu, D. Study on Landslide Risk Assessment Based on Multi-Source Spatio-Temporal Data. Ph.D. Thesis, China University of Geosciences, Wuhan, China, 2017. [Google Scholar]
- Zhao, C.; Qian, L. Comparative Study of Supervised and Unsupervised Classification in Remote Sensing Image. J. Henan Univ. (Nat. Sci.)
**2004**, 34, 90–93. [Google Scholar] - Zhou, P.; Cheng, J.; Yao, X.; Han, J. Machine Learning Paradigm in High Resolution remote Sensing Image interpretation. J. Remote Sens.
**2021**, 25, 182–197. [Google Scholar] [CrossRef] - Li, A.S.; Shi, W.X.; Zhang, Y.K. Genesis Analysis of the Landslide in Lanzhou and the Prevention and Control. Subgrade Eng.
**2018**, 6, 226–229. [Google Scholar] - Yan, Y.; Dong, X.; Li, Y. The comparative study of Remote Sensing Image Supervised classification methods based on ENVI. Beijing Surv. Mapp.
**2011**, 3, 14–16. [Google Scholar] - Wang, Y.; Wen, B. Study on geological hazard risk of landslides in Lanzhou. Chin. Geol.
**2011**, 38, 1593–1598. [Google Scholar] - Wu, W.; Feng, X.; Wang, Z.; Wang, W. Landslide and debris flow disaster and its prevention in Lanzhou. J. Xi’an Inst. Geosci.
**1996**, 18, 43–50. [Google Scholar] - Zhao, Z.; Wang, J.; Mao, X.; Ma, W.; Lu, W.; He, Y.; Gao, X. Landslide susceptibility assessment method coupled with multi-dimensional CNN. J. Wuhan Univ. (Inf. Sci. Ed.)
**2023**. [Google Scholar] - He, Y.; Huang, Y.; Li, H.; Liu, Q.; Liu, G.; Zhou, Z.; Zhang, C. Random forest land cover classification based on Sentinel-2A image feature optimization. Resour. Sci.
**2019**, 41, 992–1001. [Google Scholar] - Revel, C.; Lonjou, V.; Marcq, S.; Desjardins, C.; Fougnie, B.; Coppolani, C.; Guilleminot, N.; Lacamp, A.; Lourme, E.; Miquel, C.; et al. Sentinel-2A and 2B absolute calibration monitoring. Eur. J. Remote Sens.
**2019**, 52, 122–137. [Google Scholar] [CrossRef] - Pham, B.T.; Tien Bui, D.; Pourghasemi, H.R.; Indra, P.; Dholakia, M. Landslide susceptibility assesssment in the Uttarakhand area (India) using GIS: A comparison study of prediction capability of naïve bayes, multilayer perceptron neural networks, and functional trees methods. Theor. Appl. Climatol.
**2017**, 128, 255–273. [Google Scholar] [CrossRef] - Luna-Alvarez, A.; Mujica-Vargas, D.; Matuz-Cruz, M.; Kinani, J.; Ramos-Díaz, E. Self-Driving through a Time-Distributed Convolutional Recurrent Neural Network; IEEE: Piscataway, NJ, USA, 2020. [Google Scholar]
- Chauhan, S.; Sharma, M.; Arora, M.K.; Gupta, N.K. Landslide susceptibility zonation through ratings derived from artificial neural network. Int. J. Appl. Earth Obs. Geoinf.
**2010**, 12, 340–350. [Google Scholar] [CrossRef] - Guo, P.; Ai, Y.; Chen, C.; Li, W.; Wang, Q.; Pei, J. Effects of plant soil types on physicochemical properties and microbial activity of artificial soil on rock slope. J. Soil Water Conserv.
**2012**, 26, 6. [Google Scholar] - Cao, P. Study on Formation Mechanism and Stability of Baige High Landslide in Eastern Tibet. Master’s Thesis, Kunming University of Science and Technology, Kunming, China, 2021. [Google Scholar]
- Pu, H. InSAR Technology Based on Improved DBN Zhouqu County landslide Susceptibility Assessment. Master’s Thesis, Lanzhou Jiaotong University, Lanzhou, China, 2022. [Google Scholar]
- Sun, C.; Wu, Z.; Lv, Z.; Yao, N.; Wei, J. Quantifying different types of urban growth and the change dynamic in Guangzhou using multi-temporal remote sensing data. Int. J. Appl. Earth Obs. Geoinf.
**2013**, 21, 409–417. [Google Scholar] [CrossRef] - Li, A.; Hou, S.; Zhou, P. Study on rainfall-induced landslides in Yucheng District, Yaan City, Sichuan Province. Chin. J. Geol. Hazards Prev.
**2007**, 18, 15–17. [Google Scholar] - Zheng, Y.; Chen, J.; Wang, C.; Cheng, T. Application of certainty coefficient and stochastic forest model to the vulnerability assessment of landslides in Mangshi, Yunnan. Geol. Sci. Technol. Bull.
**2020**, 39, 131–144. [Google Scholar] - Tu, E.; Yang, J. An overview of semi-supervised learning theory and its research progress. J. Shanghai Jiaotong Univ.
**2018**, 52, 1280–1291. [Google Scholar] - Li, H.; Sun, Z.; Wu, Y.; Song, Y. Semi-supervised point cloud segmentation using self-training with label confidence prediction. Neurocomputing
**2021**, 437, 227–237. [Google Scholar] [CrossRef] - Zoph, B.; Ghiasi, G.; Lin, T.Y.; Yin, C.; Liu, H.; Cubuk, E.; Le, Q. Rethinking pre-training and self-training. Adv. Neural Inf. Process. Syst.
**2020**, 33, 3833–3845. [Google Scholar] - Ning, X.; Wang, X.; Xu, S.; Cai, W.; Zhang, L.; Yu, L.; Li, W. A review of research on co-training. Concurr. Comput. Pract. Exp.
**2021**, 35, e6276. [Google Scholar] [CrossRef] - Tran, H.Q.; Ha, C. Reducing the burden of data collection in a fingerprinting-based VLP system using a hybrid of improved co-training semi-supervised regression and adaptive boosting algorithms. Opt. Commun.
**2021**, 488, 126857. [Google Scholar] [CrossRef] - Wang, K.; Gou, C.; Duan, Y.; Lin, Y.; Zheng, X.; Wang, F. Research progress and prospect of generative countermeasure network GAN. J. Autom.
**2017**, 43, 321–332. [Google Scholar] - Lv, H.; Yu, L.; Zhou, X.; Deng, X. A review of semi-supervised deep learning image classification methods. Comput. Sci. Explor.
**2021**, 15, 1038–1048. [Google Scholar] - Goodfellow, I.; Pouget-Abadie, J.; Mirza, M.; Bing, X.; Warde-Farley, D.; Ozair, S.; Courville, A.; Bengio, Y. Generative adversarial networks. Commun. ACM
**2020**, 11, 139–144. [Google Scholar] [CrossRef] - Lin, Y.L.; Dai, X.Y.; Li, L.; Wang, X.; Wang, F. The New Frontier of AI Research: Generative Adversarial Networks. Acta Autom. Sin.
**2018**, 44, 775–792. [Google Scholar] - Liang, J.; Wei, J. Summary of generating countermeasure network GAN. Comput. Sci. Explor.
**2020**, 14, 1–17. [Google Scholar] - Zhou, T.; Liu, W.; Zhou, C.; Chen, L. GAN-Based Semi-Supervised for Imbalanced Data Classification; IEEE: Piscataway, NJ, USA, 2018. [Google Scholar]
- Miao, X.; Wu, Y.; Wang, J.; Mao, Y.; Yin, X.; Jian, W. Generative Semi-Supervised Learning for Multivariate Time Series Imputation. Proceedings of the. AAAI Conf. Artif. Intell.
**2021**, 35, 8983–8991. [Google Scholar] [CrossRef] - Ren, S.; He, K.; Girshick, R.; Sun, J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans. Pattern Anal. Mach. Intell.
**2017**, 39, 1137–1149. [Google Scholar] [CrossRef] [PubMed] - Kyriou, A.; Nikolakopoulos, K. Landslide mapping using optical and radar data: A case study from Aminteo, Western Macedonia Greece. Eur. J. Remote Sens.
**2020**, 53 (Suppl. 2), 17–27. [Google Scholar] [CrossRef] - Dille, A.; Kervyn, F.; Handwerger, A.L.; d‘Oreye, N.; Derauw, D.; Mugaruka, B.; Samsonov, S.; Malet, J.; Kervyn, M.; Dewitte, O. When image correlation is needed; unravelling the complex dynamics of a slow-moving landslide in the tropics with dense radar and optical time series. Remote Sens. Environ.
**2021**, 258, 112402. [Google Scholar] [CrossRef] - Wang, Y.; Fang, Z.; Wang, M.; Peng, L.; Hong, H. Comparative study of landslide susceptibility mapping with different recurrent neural networks. Comput. Geosci.
**2020**, 138, 104445. [Google Scholar] [CrossRef] - Chen, W.; Peng, J.; Hong, H.; Shahabi, H.; Pradhan, B.; Liu, J.; Zhu, A.; Pei, X.; Duan, Z. Landslide susceptibility modelling using GIS—Based machine learning techniques for Chongren county, Jiangxi Province, China. Sci. Total Environ.
**2018**, 626, 1121–1135. [Google Scholar] [CrossRef] [PubMed] - He, Y.; Yao, S.; Yang, W.; Yan, H.; Zhang, L.; Wen, Z.; Zhang, Y.; Liu, T. An Extraction Method for Glacial Lakes Based on Landsat-8 Imagery Using an Improved U-Net Network. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.
**2021**, 14, 6544–6558. [Google Scholar] [CrossRef] - Jiri, W.J.; Tian, H.; Han, J. Vulnerability evaluation of earthquake landslides based on different machine learning algorithms--taking Ludian earthquake as an example. J. Kunming Univ. Sci. Technol. (Nat. Sci. Ed.)
**2022**, 47, 47–56. [Google Scholar]

**Figure 2.**Influencing factors for landslides. (

**a**) Influencing factors of elevation. (

**b**) Influencing factors of slope. (

**c**) Influencing factors of aspect. (

**d**) Influencing factors of lithology. (

**e**) Influencing factors of distance to faults. (

**f**) Influencing factors of distance to rivers. (

**g**) Influencing factors of land use. (

**h**) Influencing factors of distance to roads. (

**i**) Influencing factors of NDVI. (

**j**) Influencing factors of average annual rainfall.

**Figure 3.**Semi-Supervised Generative Adversarial Network Structure. The red arrow represents the input and discriminant result of the real sample, and the green arrow represents the input and discriminant result of the generated sample. The yellow represents the factor, the red and orange are the labels. Grey represents random noise data vector and input data vector respectively.

**Figure 5.**Comparison of Discriminator Training Process. (

**a**) Sample set of landslide impact factors. (

**b**) Sentinel-2A optical remote sensing sample set. (

**c**) Joint sample set.

**Figure 8.**Comparison of accurate landslide recognition results in the main urban area of Lanzhou City: (

**a**) GAN; (

**b**) SSGAN. Pink represents the accurate landslide recognition results by GAN, green represents the accurate landslide recognition results by SSGAN.

Sensor | Wave Band | Central Wavelength (μm) | Resolution Ratio (m) |
---|---|---|---|

MSI | Band 1—super blue (coastal and aerosol) | 0.443 | 60 |

MSI | Band 2—blue | 0.490 | 10 |

MSI | Band 3—green | 0.560 | 10 |

MSI | Band 4—red | 0.665 | 10 |

MSI | Band 5—visible and near-infrared light | 0.705 | 20 |

MSI | Band 6—visible and near-infrared light | 0.740 | 20 |

MSI | Band 7—visible and near-infrared light | 0.783 | 20 |

MSI | Band 8—near-infrared light | 0.842 | 10 |

MSI | Band 8A—visible and near-infrared light | 0.865 | 20 |

MSI | Band 9—shortwave infrared-vapor | 0.945 | 60 |

MSI | Band 10—shortwave infrared–cirrus cloud | 1.375 | 60 |

MSI | Band 11—shortwave infrared | 1.610 | 20 |

MSI | Band 12—shortwave infrared | 2.190 | 20 |

Data Name | Data Source | URL | Influencing Factor | Resolution |
---|---|---|---|---|

SRTM DEM | United States Geological Survey | http://earthexplorer.usgs.gov (26 June 2023) | Elevation | 30 m |

Slope | ||||

Aspect | ||||

Land Use | Tsinghua University | http://data.ess.tsinghua.edu.cn/ (11 September 2023) | Land use | 30 m |

Rainfall | Geospatial Remote Sensing Ecology Network | http://www.gisrs.cn/ (6 March 2024) | Average annual rainfall | 50 m |

Normalized Difference Vegetation Index | National Ecological Data Center Resource Sharing Platform | http://www.nesdc.org.cn (18 January 2024) | NDVI | 10 m |

Lithology and Faults | United States Geological Survey | https://www.cgs.gov.cn/ (20 December 2023) | Lithology | 1:4,000,000 |

Distance to faults | 30 m | |||

Rivers and Roads | Lanzhou Natural Resources Bureau | http://zrzyj.lanzhou.gov.cn/ (24 October 2023) | Distance to rivers | 30 m |

Distance to roads | 30 m |

Confusion Matrix | Predicted Value | ||
---|---|---|---|

Positive Example (+) | Counter Example (−) | ||

True value | Positive example (+) | True Positive TP | False Negative FN |

Counter example (−) | False Positive FP | True Negative TN |

Precision | Recall | F1 Score | Kappa Coefficient | MIoU | |
---|---|---|---|---|---|

Model a (GAN) | 0.795 | 0.961 | 0.860 | 0.858 | 0.887 |

Model b (SSGAN) | 0.829 | 0.952 | 0.879 | 0.878 | 0.899 |

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |

© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Lu, W.; Zhao, Z.; Mao, X.; Cheng, Y.
An Accurate Recognition Method for Landslides Based on a Semi-Supervised Generative Adversarial Network: A Case Study in Lanzhou City. *Appl. Sci.* **2024**, *14*, 5084.
https://doi.org/10.3390/app14125084

**AMA Style**

Lu W, Zhao Z, Mao X, Cheng Y.
An Accurate Recognition Method for Landslides Based on a Semi-Supervised Generative Adversarial Network: A Case Study in Lanzhou City. *Applied Sciences*. 2024; 14(12):5084.
https://doi.org/10.3390/app14125084

**Chicago/Turabian Style**

Lu, Wenjuan, Zhan’ao Zhao, Xi Mao, and Yao Cheng.
2024. "An Accurate Recognition Method for Landslides Based on a Semi-Supervised Generative Adversarial Network: A Case Study in Lanzhou City" *Applied Sciences* 14, no. 12: 5084.
https://doi.org/10.3390/app14125084