Feature Extraction with Multi-fractal Spectrum for Coal and Gangue Recognition Based on Texture Energy Field
Tóm tắt
Feature extraction is an important part for coal and gangue recognition, which has direct impact on accuracy of recognition. However, the existing feature extraction methods for coal and gangue are not ideal, and so a feature extraction method with multi-fractal is proposed based on energy field normalization for target recognition of coal and gangue in this paper. In the method, the concept of target energy field is established based on 3D grey surface, and the normalized target energy is calculated by using pixels. Then, after analysis of the feature extraction process, a feature extraction algorithm with multi-fractal is proposed based on energy field, in which 3D grey surface is divided by different grid forms, and the pixels in grid are counted to obtain the probability density distribution matrix of pixels. The results of multiple feature extraction is observed visually from probability density distribution, spatial feature distribution, and multi-fractal spectrum to illustrate the measurability of the method for target textures, which is the quantitative attribute of feature extraction. In the experiment, this method is used to quantitatively measure grey texture, and the effectiveness of measured features in coal and gangue recognition is compared with other methods. The experiment results show that the method can achieve effective quantitative measurement for coal and gangue texture, and the recognition accuracy is higher than other methods.
Tài liệu tham khảo
Baravalle, R. G., Delrieux, C. A., & Gomez, J. C. (2015). Multifractal characterisation and classification of bread crumb digital images. Eurasip. Journal on Image and Video Processing, 9, 1–10.
Cao, W. L., Shi, Z. K., & Feng, J. H. (2007). Traffic image classification method based on fractal dimension. In 2006 5th IEEE international conference on cognitive informatics, pp. 903–907. https://doi.org/10.1109/COGINF.2006.365612.
Cao, X. G., Li, Y., & Wang, P. (2020). Current status and prospects of research on coal gangue identification methods. Industrial and Mining Automation, 46(1), 38–43.
Cherouat, S., Soltani, F., & Schmitt, F. (2015). Using fractal dimension to target detection in bistatic SAR data. Signal, Image and Video Processing, 9(2), 365–371.
Dai, L., Wang, K. K., & Zhu, Y. (2016). Environment image recognition based on multifractal and improved bp algorithm. Electronic Design Engineering, 24(17), 167–170.
Don, S., Chung, D., Revathy, K., Eunmi, C., & Dugki, M. (2009). A neural network approach to mammogram image classification using fractal features. In 2009 IEEE International conference on intelligent computing and intelligent systems, pp. 444–447. https://doi.org/10.1109/ICICISYS.2009.5357653.
Dong, L. Y., Shan, R., Liu, H. M., Yu, D. S., & Du, K. (2021). Research on sunken ship recognition method of side-scan sonar image based on fractal texture feature. Marine Geology and Quaternary Geology, 41(4), 232–239.
Du, G., & Yeo, T. S. (2002). A novel multifractal estimation method and its application to remote image segmentation. IEEE Transactions on Geoscience and Remote Sensing, 40(4), 980–982.
Fan, Z., Chen, N. J., & Huang, Y. L. (2021). Coal gangue recognition based on support vector machine and multiple features. Journal of Jinan University: Natural Science Edition, 35(3), 277–284.
Femmam, S. (2015). Texture classification approach based on 2D multifractal analysis. SPIE Newsroom. https://doi.org/10.1117/2.1201503.005806
Fu, C. C., Lu, F. L., & Zhang, G. Y. (2020). Discrimination analysis of coal and gangue using multifractal properties of optical texture. International Journal of Coal Preparation and Utilization, 42(7), 1–13.
Gerardo, D. M., Alessio, D. S., & Daniele, R. (2018a). Fractal-based local range slope estimation from single sar image with applications to sar despeckling and topographic mapping. Remote Sensing, 10(8), 1294–1302.
Gerardo, D. M., Antonio, I., Daniele, R., Giuseppe, R., & Ivana, Z. (2018b). The role of resolution in the estimation of fractal dimension maps from sar data. Remote Sensing, 10(2), 9–20.
Grassberger, P. (1983). Generalized dimensions of strange attractors. Physics Letters A, 97(6), 227–230.
Jiang, S., Wang, F., Shen, L. M., Li, G. P., & Wang, L. (2017). Extracting sensitive spectrum bands of rapeseed using multiscale multifractal detrended fluctuation analysis. Journal of Applied Physics, 121(10), 104–107.
Jin, C., Huang, H., & Liu, K. (2010). Medical image segmentation method based on multifractal. China Tissue Engineering Research and Clinical Rehabilitation, 14(9), 1535–1538.
Kai, L., Xi, Z., & Chen, Y. (2018). Extraction of coal and gangue geometric features with multifractal detrending fluctuation analysis. Applied Sciences, 8(3), 2–15.
Lai, K. X., Chen, L., Zhou, W. S., Yu, K., & He, T. (2016a). Research on extraction of ring gear fractal features by an improved differential box dimension method. Journal of Hubei University of Technology, 31(2), 5–8.
Lai, K. X., Li, C. C., Chen, L., & He, T. (2016b). Research on the defect recognition of synchronizer gear ring based on fractal feature. Journal of Hubei University of Technology, 31(1), 8–11.
Li, N., & Gong, X. Y. (2021). An image preprocessing model of coal and gangue in high dust and low light conditions based on the joint enhancement algorithm. Computational Intelligence and Neuroscience, 2021, 2436486.
Li, N., Gong, X. Y., & Jia, P. T. (2022a). Segmentation method for low-quality images of coal and gangue based on Retinex and local texture features with multifractal. Journal of Electronic Imaging, 31(6), 061820.
Li, N., Xue, J. M., & Gao, S. (2022b). Feature extraction method CNDFA for target contour of coal and gangue based on multifractal. Journal of Electronic Imaging, 31(4), 041217.
Li, P. F., Zhao, T. H., Zhang, X. B., Mei, S., Yan, Z. H., & Qin, K. (2015). Fractal research of remote sensing linear faults in Shandong peninsula. Marine Geology & Quaternary Geology, 35(4), 105–112.
Liu, M., Wang, P., Chen, S., & Zhang, D. (2019). The classification of inertinite macerals in coal based on the multifractal spectrum method. Applied Sciences, 9(24), 5509.
Liu, S., Zheng, P., & Cheng, X. (2017). A novel fast fractal image compression method based on distance clustering in high dimensional sphere surface. Fractals, 25(4), 17400047.
Lopes, R., & Betrouni, N. (2009). Fractal and multifractal analysis: A review. Medical Image Analysis, 13(4), 634–649.
Luo, J., Zi, C., Zhang, J., & Liu, Y. (2017). Pedestrian detection based on multifractal spectrum. Journal of Tianjin University of Technology, 36(2), 59–63.
Moctezuma, R. E., & Gonzlez-Gutirrez, J. (2020). Multifractal structure in sand drawings. Fractals- Complex Geometry Patterns and Scaling in Nature and Society, 28(1), 1–13.
Palanisamy, R., Swaminathan, R., & Sundar, S. (2019). Differentiation of EMCI in SMR images using segmented brainstem multifractal texture measures. Electronics Letters, 55(23), 1213–1214.
Popovic, N., Lipovac, M., Radunovic, M., Ugarte, J., Isusquiza, E., & Beristain, A. (2019). Fractal characterization of retinal microvascular network morphology during diabetic retinopathy progression. Microcirculation, 26(4), 1–12.
Potapov, A., Kuznetsov, V., & Pototskii, A. (2021). New class of topological textural multifractal descriptors and their application for processing low-contrast radar and optical images. Journal of Communications Technology and Electronics, 66(5), 581–590.
Shi, W., Zou, R. B., Wang, F., & Su, L. (2014). Image segmentation of rape pests and diseases based on multifractal. Journal of Hunan Agricultural University: Natural Science Edition, 40(5), 556–560.
Silva, P., & Florindo, J. (2021). Fractal measures of image local features: an application to texture recognition. Multimedia Tools and Applications, 8, 1–17.
Tan, C. (2017). Research on coal gangue recognition and separation technology based on image processing technology. Taiyuan University of Technology, 2–13.
Tang, H., Liu, H., Xiao, W., & Sebe, N. (2019). Fast and robust dynamic hand gesture recognition via key frames extraction and feature fusion. Neurocomputing, 331(FEB.28), 424–433.
Tarquis, A., Platonov, A., Matulka, A., Grau, J., Sekula, E., & Diez, M. (2013). Application of multifractal analysis to the study of sar features and oil spills on the ocean surface. Nonlinear Processes in Geophysics, 21(2), 439–450.
Tian, W. Q., Cheng, H. H., Liu, Q. L., Pzhang, P., & Li, J. (2015). Remote sensing image fusion detection method based on multifractal. Journal of Weapon Equipment Engineering, 36(003), 135–137.
Wang, J., Li, L., & Yang, S. (2018). Experimental study on extracting grayscale and texture features of coal gangue images under different illuminances. Journal of Coal Science, 43(11), 3051–3061.
Zeng, H. L. (2015). Research on coal gangue recognition technology based on image processing. North China University of Technology, 3–14.
Zhang, L., Sui, Y. P., Wang, H. S., Hao, S. K., & Zhang, N. B. (2022). Image feature extraction and recognition model construction of coal and gangue based on image processing technology. Scientific Report, 12(1), 20983.
Zhang, X., Zhao, J. M., Ni, X. L., Li, H. P., & Song, W. X. (2019). Fault diagnosis of rolling bearing based on multifractal descent algorithm and improved k-means clustering. Journal of Beijing University of Technology, 39(5), 473–479.
Zhao, L. J., Han, L. G., Zhang, H. N., Liu, Z. F., Gao, F., Yang, S. J., & Wang, Y. D. (2023). Study on recognition of coal and gangue based on multimode feature and image fusion. PLoS ONE, 18(2), 0281397.