Comparison of deep learning algorithms for geochemical anomaly identification
LI Mu-Si1,2,3(), CHEN Li-Rong2,3(), XIE Fei2, GU Lan-Ding2, WU Xiao-Dong2, MA Fen2, YIN Zhao-Feng2
1. Chinese Academy of Geological Sciences, Beijing 100037,China 2. Development and Research Center, China Geological Survey, Beijing 100037,China 3. School of Earth Sciences and Resources, China University of Geosciences (Beijing), Beijing 100083,China
There is a lack of selection bases in the geochemical anomaly identification and the reconstruction of the geochemical background conforming to the metallogenic distribution using deep learning algorithms with different network structures. Given this, based on the 1∶200 000 stream sediment data of the copper-zinc-silver metallogenic area in southwestern Fujian Province, this study extracted the combined structural characteristics, spatial distribution characteristics, and mixed characteristics of multiple elements in the samples using three unsupervised deep learning models, i.e., AE, MCAE, and FCAE. Then, these characteristics were used to reconstruct the geochemical background and simulate the metallogenic distribution. The results show that the anomaly areas delineated by the FCAE model were the most consistent with the known copper ore occurrences, followed by the MCAE and AE models. The FCAE, MCAE, and AE models had an area under the curve (AUC) score of 0.80, 0.78, and 0.61, respectively. Moreover, the FCAE and AE models were not sensitive to the change in the convolution window size. These results indicate that when deep learning algorithms are constructed for geochemical anomaly identification, the algorithms based on the extraction of spatial distribution characteristics or mixed characteristics perform well, and those based on the extraction of combined structural characteristics or mixed characteristics have a strong anti-interference ability for the noise caused by the change or inconsistency of the spatial observation scale. This study provides some effective selection bases for constructing geochemical anomaly identification models based on deep learning algorithms.
Guo K. The study of non-linear of complex geology land form identification of the multi-dimensioned geochemistry anomaly[D]. Chengdu: Chengdu University of Technology, 2005:12.
[2]
Tobler W. On the first law of geography: A reply[J]. Annals of the Association of American Geographers, 2004, 94(2): 304-310.
doi: 10.1111/j.1467-8306.2004.09402009.x
[3]
Zuo R G, Xiong Y H, Wang J, et al. Deep learning and its application in geochemical mapping[J]. Earth-Science Reviews, 2019, 192: 1-14.
doi: 10.1016/j.earscirev.2019.02.023
[4]
Zuo R G. Machine learning of mineralization-related geochemical anomalies: A review of potential methods[J]. Natural Resources Research, 2017, 26(4): 457-464.
doi: 10.1007/s11053-017-9345-4
Liu Y P, Zhu L X, Zhou Y Z. Application of convolutional neural network in prospecting prediction of ore deposits: Taking the Zhaojikou Pb-Zn ore deposit in Anhui Province as a case[J]. Acta Petrologica Sinica, 2018, 34(11): 3217-3224.
Cai H H, Zhu W, Li Z X, et al. Prediction method of tungsten-molybdenum prospecting target area based on deep learning[J]. Journal of Geo-information Science, 2019, 21(6):928-936.
Chen L R. Multivariate geochemical anomaly recognition using spatial constrained autoencoders[D]. Wuhan: China University of Geosciences(Wuhan), 2019:79.
[8]
Chen L R, Guan Q F, Xiong Y H, et al. A spatially constrained multi-autoencoder approach for multivariate geochemical anomaly recognition[J]. Computers & geosciences, 2019, 125:43-54.
doi: 10.1016/j.cageo.2019.01.016
[9]
Chen L R, Guan Q F, Feng B, et al. A multi-convolutional autoencoder approach to multivariate geochemical anomaly recognition[J]. Minerals, 2019, 9(5):270.
doi: 10.3390/min9050270
[10]
Guan Q F, Ren S L, Chen L R, et al. A spatial-compositional feature fusion convolutional autoencoder for multivariate geochemical anomaly recognition[J]. Computers and Geosciences, 2021(1):104890.
Gao Y. Mineral prospecting information mining and mapping mineral prospectivity for copper polymetallic mineralization in southwest Fujian Province[D]. Wuhan: China University of Geosciences(Wuhan), 2019:30.
Zhang C G, Chen R S, Huang C Q, et al. Study on the geological background of mineralization and mineralization pattern of Wuyishan mineralization zone[M]. Beijing: Geological Publishing House, 2014: 60.
Liu C M, Hu S Q, Ma S M, et al. The role of the phase state of metallogenic elements in the recognition of geochemical anomalies[J]. Geophysical and Geochemical Exploration, 2013, 37(6):1049-1055.
Zheng Z Y, Zhao Q Y, Li S X, et al. Comparison of two machine learning algorithms for geochemical anomaly detection[J]. Global Geology, 2018, 37(4): 1288-1294.
[15]
Rumelhart D E, Hinton G E, Williams R J. Learning representations by back propagating errors[J]. Nature, 1986, 323(6088): 533-536.
doi: 10.1038/323533a0
Deng J F, Zhang X L. Deep learning algorithm optimization based on combination of auto-encoders[J]. Journal of Computer Applications, 2016, 36(3): 697-702.
doi: 10.11772/j.issn.1001-9081.2016.03.697
Fei Y, Miao Q Y, Liu X J. Recommendation system attack detection method based on convolutional autoencoder[J]. Journal of Chinese Computer Systems, 2021, 42(5): 1088-1092.
Song X X. High resolution remote sensing image classification based on stacked autoencoder[J]. Geomatics & Spatial Information Technology, 2021, 44(5):128-131.
[19]
张扬. 基于卷积自编码器的异常事件检测研究[D]. 杭州: 浙江大学, 2018:10.
[19]
Zhang Y. Anomaly detection based on convolutional autoencoder[D]. Hangzhou: Zhejiang University, 2018:10.
[20]
Chen K, Seuret M, Liwicki M, et al. Page segmentation of historical document images with convolutional autoencoders[C]// 2015 13th International Conference on Document Analysis and Recognition (ICDAR), IEEE, 2015: 1011-1015.
Song H, Gao Y, Chen W, et al. Seismic noise suppression based on convolutional denoising autoencoders[J]. Oil Geophysical Prospecting, 2020, 55(6): 1210-1219.
Jiang J S, Ren H R, Li H Y. Seismic data processing based on convolutional autoencoder[J]. Journal of Zhejiang University:Engineering Science, 2020, 54(5): 978-984.
[23]
An J, Cho S. Variational autoencoder based anomaly detection using reconstruction probability[J]. Special Lecture on IE, 2015, 2(1): 1-18.
[24]
Xiong Y H, Zuo R G. Recognition of geochemical anomalies using a deep autoencoder network[J]. Computers and Geosciences, 2016, 86: 75-82.
doi: 10.1016/j.cageo.2015.10.006
[25]
Valentine A P, Trampert J. Data space reduction, quality assessment and searching of seismograms: Autoencoder networks for waveform data[J]. Geophysical Journal International, 2012, 189(2): 1183-1202.
doi: 10.1111/j.1365-246X.2012.05429.x
[26]
Fawcett T. An introduction to ROC analysis[J]. Pattern recognition letters, 2006, 27(8): 861-874.
doi: 10.1016/j.patrec.2005.10.010
[27]
Benesty J, Chen J, Huang Y, et al. Pearson correlation coefficient [G]// Noise reduction in speech processing. Berlin, Heidelberg: Springer, 2009: 1-4.
[28]
Chen Y L, Lu L J, Li X B. Application of continuous restricted Boltzmann machine to identify multivariate geochemical anomaly[J]. Journal of Geochemical Exploration, 2014, 140: 56-63.
doi: 10.1016/j.gexplo.2014.02.013
[29]
Zhou J, Cui G Q, Hu S D, et al. Graph neural networks: A review of methods and applications[J]. AI Open, 2020, 1∶ 57-81.
Chen Z J, Cheng Q M, Chen J G. Comparison of different models for anomaly recognition of geochemical data by using sample ranking method[J]. Earth Science:Journal of China University of Geosciences, 2009, 34(2):353-364.
doi: 10.3799/dqkx.2009.038
