|
|
|
|
|
|
| Mineral Spectra Classification Based on One-Dimensional Dilated Convolutional Neural Network |
| TIAN Qing-lin1, GUO Bang-jie1, YE Fa-wang1, LI Yao2, LIU Peng-fei1, CHEN Xue-jiao1 |
1. National Key Laboratory of Remote Sensing Information and Image Analysis Technology, Beijing Research Institute of Uranium Geology, Beijing 100029, China
2. Zachry Department of Civil and Environmental Engineering, Texas A&M University, Texas 77843, USA
|
|
|
|
|
Abstract The spectrum is a comprehensive reflection of the mineral’s physical chemistry characteristics, composition and structure, which has been used in mineral and rock identification. The traditional classification methods of the mineral spectrum require complex spectral pretreatment, and then some spectral features are analyzed by different methods to achieve the goal of fine classification. However, the pretreatment may cause a partial loss of the spectral information and reduce the classification accuracy. Besides, the operation process is complex, so the efficiency is low, making it difficult to cope with the growing demand for big data processing. Therefore, it is important to establish an accurate, efficient and automatic classification model for the mineral spectrum. As one of the widely used deep learning models, the convolutional neural network extracts data features layer by layer and combines them to form higher-level semantic information. It has a strong capability of model formulation and great potential for the analysis of spectral data. This paper proposes a novel mineral spectrum classification method based on a one-dimensional dilated convolutional neural network (1D-DCNN). The DCNN is used for spectral feature extraction. The backpropagation algorithm combined with the random gradient descent optimizer is used to adjust the model’s parameters, then output the classification result, which implements the end-to-end discrimination of mineral species. The 1D-DCNN includes one input layer, three dilated convolution layers, two pooling layers, two full connection layers and one output layer. It uses cross-entropy as the loss function, and dilated convolution is introduced to enlarge the receptive field of filters effectively avoid the loss of spectral feature details. The spectrum of four different minerals, muscovite, dolomite, calcite and kaolinite, are collected, and the data are augmented by way of adding noise to construct sufficient spectral samples, which are used for model training and testing. Then, we explore the impacts of different model parameters, such as the convolution type and the number of iterations, and then compare the proposed model with the traditional mineral spectrum classification methods to evaluate its performance. Experimental results indicate that the 1D-DCNN model can quickly and accurately classify mineral spectrum with the accuracy of 99.32%, which is superior to the backpropagation (BP) algorithm and support vector machine (SVM) methods, and it shows that the proposed method can fully learn mineral spectral features and implement a fine classification result, with good robustness and scalability. The proposed method can apply further to the spectra classification in coal, oil-gas, lunar soil and other fields.
|
|
Received: 2021-02-26
Accepted: 2021-06-15
|
|
|
|
[1] SUN Yu，ZHAO Ying-jun，LI Han-bo，et al(孙 雨，赵英俊，李瀚波，等). Acta Geologica Sinica(地质学报)，2017，89(1)：196.
[2] YU Dong-hang，ZHANG Bao-ming，ZHAO Chuan，et al(余东行，张保明，赵 传，等). Journal of Remote Sensing(遥感学报)，2020，24(6)：717.
[3] Jia G M, Cheng F Y, Yang J F，et al. Chinese Journal of Aeronautics，2018，31(12)：2280.
[4] Yu M X, Sun Y C, Zhu B F，et al. Neurocomputing，2020，378：270.
[5] Sainath T N，Kingsbury B，Saon G，et al. Neural Networks，2015，64：39.
[6] Gu J，Wang Z，Kuen J，et al. Pattern Rocognition，2018, 77: 354.
[7] HE Dong-yuan，LIU Wei，CAO Shuo，et al(何东远，刘 伟，曹 硕，等). Journal of Beijing Normal University·Natural Science(北京师范大学学报·自然科学版)，2020，56(1)：37.
[8] ZHAO Yong，RONG Kang，TAN Ai-ling(赵 勇，荣 康，谈爱玲). Spectroscopy and Spectral Analysis(光谱学与光谱分析)，2019，39(12)：3755.
[9] HU Li，SHAN Rui，WANG Fang，et al(胡 丽，单 锐，王 芳，等). Laser & Optoelectronics Progresss(激光与光电子学进展)，2020，57(12)：122803.
[10] Chen L C，Papandreou G，Kokkinos I，et al. Deaplab: Semantic Image Sengmentation With Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs, IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018.
|
| [1] |
WANG Yu-chen1, 2, KONG Ling-qin1, 2, 3*, ZHAO Yue-jin1, 2, 3, DONG Li-quan1, 2, 3*, LIU Ming1, 2, 3, HUI Mei1, 2. Hyperspectral Reconstruction From RGB Images for Tissue Oxygen
Saturation Assessment[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(10): 3193-3201. |
| [2] |
YE Wen-chao1, LUO Shui-yang1, LI Jin-hao1, LI Zhao-rong1, FAN Zhi-wen1, XU Hai-tao1, ZHAO Jing1, LAN Yu-bin1, 2, DENG Hai-dong1*, LONG Yong-bing1, 2, 3*. Research on Classification Method of Hybrid Rice Seeds Based on the Fusion of Near-Infrared Spectra and Images[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(09): 2935-2941. |
| [3] |
FAN Ya-wen, LIU Yan-ping*, QIU Bo, JIANG Xia, WANG Lin-qian, WANG Kun. Research on Spectral Classification of Stellar Subtypes Based on
SSTransformer[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(08): 2523-2528. |
| [4] |
TANG Ting, PAN Xin*, LUO Xiao-ling, GAO Xiao-jing. Fusion of ConvLSTM and Multi-Attention Mechanism Network for
Hyperspectral Image Classification[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(08): 2608-2616. |
| [5] |
MAO Yi-lin1, LI He1, WANG Yu1, FAN Kai1, SUN Li-tao2, WANG Hui3, SONG Da-peng3, SHEN Jia-zhi2*, DING Zhao-tang1, 2*. Quantitative Judgment of Freezing Injury of Tea Leaves Based on Hyperspectral Imaging[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(07): 2266-2271. |
| [6] |
JIANG Xia*, QIU Bo, WANG Lin-qian, GUO Xiao-yu. Automatic Classification Method of Star Spectra Based on
Semi-Supervised Mode[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(06): 1875-1880. |
| [7] |
LIU Si-qi1, FENG Guo-hong1*, TANG Jie2, REN Jia-qi1. Research on Identification of Wood Species by Mid-Infrared Spectroscopy Based on CA-SDP-DenseNet[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(03): 814-822. |
| [8] |
LENG Si-yu1, 2, QIAO Jia-hui1, WANG Lian-qing3, WANG Jun1, 2*, ZOU Liang1. Rapid Qualitative Analysis of Wool Content Based on Improved
U-Net++ and Near-Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(01): 303-309. |
| [9] |
FU Peng-you1, 2, WEN Yue2, ZHANG Yu-ke3, LI Ling-qiao1*, YANG Hui-hua1, 2*. Deep Learning Modelling and Model Transfer for Near-Infrared Spectroscopy Quantitative Analysis[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(01): 310-319. |
| [10] |
CUI De-jian1, LIU Yang-yang1, XIA Yuan-tian1, JIA Wei-e1, LIAN Zheng-xing2, LI Lin1*. Non-Destructive Detection of Egg Fertilization Status Based on Hyperspectral Diffuse Reflectance[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(12): 3685-3691. |
| [11] |
NI Shuang, WEN Jia-xing, ZHOU Min-jie, HUANG Jing-lin, LE Wei, CHEN Guo, HE Zhi-bing, LI Bo, ZHAO Song-nan, ZHAO Zong-qing, DU Kai*. Theoretical Study on Raman Characteristic Peaks of Coronavirus Spike Protein Based on Deep Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(09): 2757-2762. |
| [12] |
ZHANG Shuai-shuai1, GUO Jun-hua1, LIU Hua-dong1, ZHANG Ying-li1, XIAO Xiang-guo2, LIANG Hai-feng1*. Design of Subwavelength Narrow Band Notch Filter Based on
Depth Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(05): 1393-1399. |
| [13] |
TAN Ai-ling1, CHU Zhen-yuan1, WANG Xiao-si1, ZHAO Yong2*. Detection of Pearl Powder Adulteration Based on Raman Spectroscopy and DCGAN Data Enhancement[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 769-775. |
| [14] |
SUN Zhi-xing, ZHAO Zhong-gai*, LIU Fei. Near-Infrared Spectral Modeling Based on Stacked Supervised Auto-Encoder[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 749-756. |
| [15] |
LIU Zhong-bao1, WANG Jie2*. Research on the Improvement of Spectra Classification Performance With the High-Performance Hybrid Deep Learning Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 699-703. |
|
|
|
|
