基于复合分形的路面抗滑随机森林评估模型

彭毅; 张政奇; 李强; 杨广伟

doi:10.21656/1000-0887.440244

基于复合分形的路面抗滑随机森林评估模型

doi: 10.21656/1000-0887.440244

彭毅^{1, 2, ,},
张政奇³,
李强⁴,
杨广伟⁴

1.
重庆交通大学交通运输学院，重庆 400074
2.
重庆交通大学交通土建工程材料国家地方联合工程研究中心，重庆 400074
3.
重庆交通大学土木工程学院，重庆 400074
4.
俄克拉荷马州立大学土木与环境工程学院，俄克拉荷马斯蒂尔沃特 74078-5013，美国

基金项目:

国家自然科学基金青年科学基金 52208425

重庆市博士后自然科学基金 cstc2021jcyj-bshX0113

中国博士后面上项目地区基金 2021M693918

详细信息

通讯作者:
彭毅(1987—)，男，博士(通讯作者. E-mail: dawsonyp@cqjtu.edu.cn)

中图分类号: U416.2
计量
- 文章访问数: 63
- HTML全文浏览量: 20
- PDF下载量: 16
- 被引次数: 0
出版历程
- 收稿日期: 2023-08-17
- 修回日期: 2023-12-05
- 刊出日期: 2024-04-01

A Random Forest Evaluation Model for Pavement Skid Resistance Based on Comprehensive Fractal

PENG Yi^{1, 2
, ,},
ZHANG Zhengqi³,
LI Qiang⁴,
YANG Guangwei⁴

1.
School of Traffic and Transportation, Chongqing Jiaotong University, Chongqing 400074, P.R.China
2.
National & Local Joint Engineering Research Center of Transportation and Civil Engineering Materials, Chongqing Jiaotong University, Chongqing 400074, P.R.China
3.
School of Civil Engineering, Chongqing Jiaotong University, Chongqing 400074, P.R.China
4.
School of Civil and Environmental Engineering, Oklahoma State University, Stillwater, Oklahoma 74078-5013, USA

摘要

摘要: 路面抗滑性能直接影响着道路交通安全，而基于路表纹理特征的路面抗滑性能评估方法目前存在着可解释性差、准确度不高的问题. 该研究使用精度为0.05 mm的便携式三维激光表面分析仪采集了185组路面纹理数据，通过动态摩擦因数测试仪获得了相应路段0~80 km/h速度范围内的路面摩擦数据，构建了综合表征路面纹理空间、横剖、深度方向复杂度的复合分形维数指标，建立了10 km/h和70 km/h速度下的路面抗滑性能随机森林评估模型. 研究结果表明：复合分形维数具备独立描述纹理复杂程度的能力，但与路面动态摩擦因数之间不存在线性关系；复合分形维数对70 km/h速度下动态摩擦因数预估的准确度为0.78，可用于评价轮胎橡胶快速滑动状态下的路面抗滑性能；复合分形指标中的空间、横剖、表层、浅层、深层剖面分形特征共同影响着路面抗滑性能，在进行路面纹理形貌评价时，应从多种空间视角下进行纹理特征综合分析.
- 道路工程 /
- 路面纹理 /
- 抗滑性能 /
- 分形维数 /
- 随机森林
Abstract: The pavement anti-skid performance directly affects road traffic safety, and the evaluation methods based on pavement texture features currently have problems of poor interpretability and low accuracy. Herein, 185 sets of pavement texture data were collected by the portable 3D laser surface analyzer with an accuracy of 0.05 mm. The pavement friction data in the speed range of 0~80 km/h of the corresponding road section were obtained with the dynamic friction coefficient tester. The comprehensive fractal dimension index representing the complexity of the pavement texture space, the cross section, and the depth direction was constructed, and the random forest evaluation model for pavement skid resistance performances at speeds of 10 km/h and 70 km/h. The results show that, the comprehensive fractal dimension has the ability to describe the complexity of texture independently, but there is no linear relationship between it and the pavement dynamic friction coefficient; the prediction accuracy of comprehensive fractal dimensions for dynamic friction coefficients at the 70 km/h speed is 0.78, which can be used to evaluate the skid resistance of pavement under the condition of rapid sliding of tire rubber; the spatial, cross-sectional, surface, shallow, and deep profile fractal features in comprehensive fractal indicators jointly affect the pavement anti-skid performances. In the evaluation of pavement texture morphology, comprehensive analysis of texture features should be conducted from multiple spatial perspectives.
- road engineering /
- pavement texture /
- skid resistance /
- fractal dimension number /
- random forest

HTML全文

图 1 三维分形示意

Figure 1. The 3D fractal schematic diagrams

下载: 全尺寸图片幻灯片

图 2 纹理横剖面图示

Figure 2. The texture cross-section diagram

下载: 全尺寸图片幻灯片

图 3 路面-轮胎有效接触区域示意

注为了解释图中的颜色，读者可以参考本文的电子网页版本，后同.

Figure 3. Schematic diagrams of the effective tire-pavement contact area

下载: 全尺寸图片幻灯片

图 4 路面不同深度横截面示意图

Figure 4. Cross section diagram of pavement at different depths

下载: 全尺寸图片幻灯片

图 5 最大纹理深度

Figure 5. The maximum texture depth

下载: 全尺寸图片幻灯片

图 6 纹理深度剖面示意图

Figure 6. Cross section schematic diagrams

下载: 全尺寸图片幻灯片

图 7 路面纹理点云采集示意

Figure 7. Schematic diagram of the pavement texture point cloud collection

下载: 全尺寸图片幻灯片

图 8 动态摩擦因数测试仪

Figure 8. The dynamic friction coefficient tester

下载: 全尺寸图片幻灯片

图 9 DFT测试曲线示例

Figure 9. Examples of DFT test curves

下载: 全尺寸图片幻灯片

图 10 路面动态摩擦测试结果

Figure 10. The pavement dynamic friction test results

下载: 全尺寸图片幻灯片

图 11 随机森林回归算法原理示意图

Figure 11. Schematic diagram of the random forest regression algorithm

下载: 全尺寸图片幻灯片

图 12 随机森林中示例决策树的部分节点

Figure 12. Some nodes of the example decision tree in the random forest

下载: 全尺寸图片幻灯片

图 13 特征箱线图

Figure 13. The characteristic boxplot

下载: 全尺寸图片幻灯片

图 14 路面抗滑性能预测结果

Figure 14. Prediction results of pavement anti-skid performances

下载: 全尺寸图片幻灯片

图 15 复合分形维数特征重要性

Figure 15. Importance of comprehensive fractal dimension number characteristics

下载: 全尺寸图片幻灯片

表 1 不同速度下动态摩擦因数相关性

Table 1. Correlation of dynamic friction coefficients at different speeds

	dynamic friction coefficient
	DFT70	DFT60	DFT50	DFT40	DFT30	DFT25	DFT20	DFT15	DFT10
DFT70	1.00	0.99	0.98	0.95	0.90	0.85	0.74	0.51	0.26
DFT60	0.99	1.00	0.99	0.97	0.93	0.89	0.78	0.55	0.29
DFT50	0.98	0.99	1.00	0.99	0.96	0.91	0.81	0.57	0.30
DFT40	0.95	0.97	0.99	1.00	0.98	0.95	0.86	0.63	0.36
DFT30	0.90	0.93	0.96	0.98	1.00	0.99	0.92	0.72	0.46
DFT25	0.85	0.89	0.91	0.95	0.99	1.00	0.97	0.8	0.56
DFT20	0.74	0.78	0.81	0.86	0.92	0.97	1.00	0.92	0.73
DFT15	0.51	0.55	0.57	0.63	0.72	0.8	0.92	1.00	0.89
DFT10	0.26	0.29	0.30	0.36	0.46	0.56	0.73	0.89	1.00

下载: 导出CSV

表 2 随机森林超参数

Table 2. Hyperparameter of the random forest

hyperparameter	meaning	explanatory note
n_estimator	number of random forest learners	the default value is 10, a model with too small n_estimator is prone to underfitting, while conversely, it requires a too big computation
d_max	maximum depth of the decision tree	default value be not limited, and adjusted based on the sample size and feature size of the data

下载: 导出CSV

表 3 复合分形维数与路面摩擦因数的组间相关性

Table 3. The intergroup correlation between the comprehensive fractal dimension number and the pavement friction coefficient

	F_3D	F_Sur	F_S	F_D	F_2D
DFT70	0.215 6	0.097 5	0.006 2	0.037 5	0.000 5
DFT10	0.079 6	0.001 4	0.058 5	0.003 7	0.045 9

下载: 导出CSV

表 4 复合分形维数组内特征相关性

Table 4. The intragroup feature correlation of comprehensive fractal dimension numbers

feature	F_3D	F_Sur	F_S	F_D	F_2D
F_3D	1.00	0.004 9	-0.23	-0.23	0.45
F_Sur	0.004 9	1.00	-0.29	-0.6	0.4
F_S	-0.23	-0.29	1.00	0.6	-0.51
F_D	-0.23	-0.6	0.6	1.00	-0.56
F_2D	0.45	0.4	-0.51	-0.56	1.00

下载: 导出CSV

表 5 随机森林模型预测评价指标

Table 5. Random forest model predictive evaluation indexes

evaluation index	10 km/h		70 km/h
evaluation index	training set	test set	training set	test set
R²	0.87	0.66	0.91	0.78
RMSE δ	0.029 6	0.043 6	0.023 1	0.040 5

下载: 导出CSV

参考文献(51)

[1]	赵雪涛. 公路路面摩擦系数检测与评价方法研究[J]. 交通世界, 2019, 23(8): 56-58. doi: 10.3969/j.issn.1006-8872.2019.08.022 ZHAO Xuetao. Research on testing and evaluation methods for friction coefficient of highway pavements[J]. Transpo World, 2019, 23(8): 56-58. (in Chinese) doi: 10.3969/j.issn.1006-8872.2019.08.022
[2]	PAN B H, CHAI H, LU B. Determination of the friction coefficients of icy pavements under different amounts of snowfall[J]. Journal of Visualized Experiments, 2023(191): e63769.
[3]	宋江春, 李明亮, 李俊, 等. 排水沥青路面抗滑性能评价指标研究[J]. 现代交通技术, 2020, 17(3): 11-16. doi: 10.3969/j.issn.1672-9889.2020.03.003 SONG Jiangchun, LI Mingliang, LI Jun, et al. Study on evaluation index of skidding performance in porous pavement[J]. Modern Transportation Technology, 2020, 17(3): 11-16. (in Chinese) doi: 10.3969/j.issn.1672-9889.2020.03.003
[4]	刘俊卿, 韩晶. 基于模糊数学理论的沥青路面结构可靠度分析[J]. 应用数学和力学, 2018, 39(9): 1081-1090. doi: 10.21656/1000-0887.380278 LIU Junqing, HAN Jing. Reliability analysis of asphalt pavement structure based on the fuzzy mathematics theory[J]. Applied Mathematics and Mechanics, 2018, 39(9): 1081-1090. (in Chinese) doi: 10.21656/1000-0887.380278
[5]	黄晓明, 郑彬双. 沥青路面抗滑性能研究现状与展望[J]. 中国公路学报, 2019, 32(4): 32-49. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201904004.htm HUANG Xiaoming, ZHENG Binshuang. Research status and progress for skid resistance performance of asphalt pavements[J]. China Journal of Highway and Transport, 2019, 32(4): 32-49. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201904004.htm
[6]	KOUCHAKI S, ROSHANI H, PROZZI J A, et al. Field investigation of relationship between pavement surface texture and friction[J]. Transportation Research Record, 2018, 2672(40): 1-13. doi: 10.1177/0361198118767415
[7]	UECKERMANN A, WANG D, OESER M. A contribution to non-contact skid resistance measurement[J]. International Journal of Pavement Engineering, 2014, 16(7): 646-659.
[8]	MEEGODA J N, GAO S Y. Evaluation of pavement skid resistance using high speed texture measurement[J]. Journal of Traffic and Transportation Engineering (English Edition), 2015, 2(6): 382-390. doi: 10.1016/j.jtte.2015.09.001
[9]	丁世海, 战友, 阳恩慧, 等. 基于高精度激光断面高程的沥青路面MTD测量[J]. 东南大学学报(自然科学版), 2020, 50(1): 137-142. https://www.cnki.com.cn/Article/CJFDTOTAL-DNDX202001018.htm DING Shihai, ZHAN You, YANG Enhui, et al. MTD measurement of asphalt pavement based on high precision laser section elevation[J]. Journal of Southeast University (Natural Science Edition), 2020, 50(1): 137-142. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-DNDX202001018.htm
[10]	AHMAD A, OMAR S, GEORGES B S, et al. Pavement friction modeling using texture measurements and pendulum skid tester[J]. Transportation Research Record, 2018, 2672(40): 1-12. doi: 10.1177/0361198118767415
[11]	LIU Q F, SHALABY A. Relating concrete pavement noise and friction to three-dimensional texture parameters[J]. International Journal of Pavement Engineering, 2017, 18(5): 450-458. doi: 10.1080/10298436.2015.1095897
[12]	HU L Q, YUN D, LIU Z Z, et al. Effect of three-dimensional macrotexture characteristics on dynamic frictional coefficient of asphalt pavement surface[J]. Construction & Building Materials, 2016, 126: 720-729.
[13]	PANAGOULI O K, KOKKALIS A G. Skid resistance and fractal structure of pavement surface[J]. Chaos, Solitons & Fractals, 1998, 9(3): 493-505.
[14]	KOKKALIS A G, PANAGOULI O K. Fractal evaluation of pavement skid resistance variations Ⅰ: surface wetting[J]. Chaos, Solitons & Fractals, 1998, 9(11): 1875-1890.
[15]	UECKERMANN A, WANG D, OESER M, et al. Calculation of skid resistance from texture measurements [J]. Journal of Traffic and Transportation Engineering (English Edition), 2015, 2(1): 3-16. doi: 10.1016/j.jtte.2015.01.001
[16]	HARTIKAINEN L, PETRY F, WESTERMANN S. Frequency-wise correlation of the power spectral density of asphalt surface roughness and tire wet friction[J]. Wear, 317(1/2): 111-119.
[17]	KANAFI M M, TUONONENA J. Top topography surface roughness power spectrum for pavement friction evaluation[J]. Tribology International, 2017, 107: 240-249. doi: 10.1016/j.triboint.2016.11.038
[18]	KOKKALIS A G, TSOHOS G H, PANAGOULI O K. Consideration of fractals potential in pavement skid resistance evaluation[J]. Journal of Transportation Engineering, 2002, 128(6): 591-595. doi: 10.1061/(ASCE)0733-947X(2002)128:6(591)
[19]	VILLANI M M, SCARPAS A, BONDT A, et al. Application of fractal analysis for measuring the effects of rubber polishing on the friction of asphalt concrete mixtures[J]. Wear, 2014, 320: 179-188. doi: 10.1016/j.wear.2014.08.013
[20]	ZHONG K, SUN M Z, LIU Z X, et al. Research on dynamic evaluation model and early warning technology of anti-sliding risk for the airport pavement[J]. Construction and Building Materials, 2020, 239: 117820. doi: 10.1016/j.conbuildmat.2019.117820
[21]	WANG K F, LI Y, ZHU Y S, et al. Research on characteristics of macrotexture for colored anti-skid coating on pavement based on fractal theory[J]. Transportation Research Record, 2022, 2676(6): 129-140. doi: 10.1177/03611981211072811
[22]	LIU C, ZHAN Y, DENG Q S, et al. An improved differential box counting method to measure fractal dimensions for pavement surface skid resistance evaluation[J]. Measurement, 2021, 178: 109376.
[23]	NAYAK S R, MISHRA J, PALAI G. Analysing roughness of surface through fractal dimension: a review[J]. Image and Vision Computing, 2019, 89: 21-34.
[24]	ISSA M A, ISSA M A, ISLAM M S, et al. Fractal dimension: a measure of fracture roughness and toughness of concrete[J]. Engineering Fracture Mechanics, 2003, 70(1): 125-137.
[25]	冉茂平, 肖旺新, 周兴林, 等. 基于三维分形维数的沥青路面抗滑性能研究[J]. 公路交通科技, 2016, 33(2): 28-32. https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK201602006.htm RAN Maoping, XIAO Wangxin, ZHOU Xinglin, et al. Research of skid resistance of asphalt pavement based on 3D fractal dimension[J]. Journal of Highway and Transportation Research and Development, 2016, 33(2): 28-32. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GLJK201602006.htm
[26]	童申家, 谢祥兵, 赵大勇. 沥青路面纹理分布的分形描述及抗滑性能评价[J]. 中国公路学报, 2016, 29(2): 1-7. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201602002.htm TONG Shenjia, XIE Xiangbing, ZHAO Dayong. Fractal description of texture distribution and evaluation of skid-resistance performance for asphalt pavement[J]. China Journal of Highway and Transport, 2016, 29(2): 1-7. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGL201602002.htm
[27]	钱振东, 薛永超, 张令刚. 沥青路面三维纹理分形维数及其抗滑性能[J]. 中南大学学报(自然科学版), 2016, 47(10): 3590-3596. https://www.cnki.com.cn/Article/CJFDTOTAL-ZNGD201610041.htm QIAN Zhendong, XUE Yongchao, ZHANG Linggang. 3-D textural fractal dimension and skid resistance of asphalt pavement[J]. Journal of Central South University (Science and Technology), 2016, 47(10): 3590-3596. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZNGD201610041.htm
[28]	LI L, WANG K C P, LI Q J. Geometric texture indicators for safety on AC pavements with 1 mm 3D laser texture data[J]. International Journal of Pavement Research and Technology, 2016, 9(1): 49-62.
[29]	DING S, WANG K C P, YANG E, et al. Influence of effective texture depth on pavement friction based on 3D texture area[J]. Construction and Building Materials, 2021, 287: 123002.
[30]	LIU C, ZHAN Y, DENG Q S, et al. An improved differential box counting method to measure fractal dimensions for pavement surface skid resistance evaluation[J]. Measurement, 2021, 178: 109376.
[31]	GUO W, CHU L, YANG L, et al. Determination of tire rubber-pavement directional coefficient of friction based on contact mechanism considerations[J]. Tribology International, 2023, 179: 1081781-10817814.
[32]	TAN Q L, XU X Y, LIANG H J. Physiological big data mining through machine learning and wireless sensor networks[J]. International Journal of Distributed Systems and Technologies, 2023, 14(2): 1-15.
[33]	KRIZHEVSKY A, SUSSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.
[34]	彭毅, 李强, 战友, 等. 基于区域三维纹理特征的路面抗滑性能评估[J]. 东南大学学报(自然科学版), 2020, 50(4): 667-676. https://www.cnki.com.cn/Article/CJFDTOTAL-DNDX202004010.htm PENG Yi, LI Qiang, ZHAN You, et al. Pavement skid resistance evaluation based on 3D areal texture characterization[J]. Journal of Southeast University (Natural Science Edition), 2020, 50(4): 667-676. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-DNDX202004010.htm
[35]	战友, 李强, 马啸天, 等. 基于宏微观纹理特征融合的路面摩擦性能预测[J]. 浙江大学学报(工学版), 2021, 55(4): 684-694. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDZC202104010.htm ZHAN You, LI Qiang, MA Xiaotian, et al. Macro and micro texture based prediction of pavement surface friction[J]. Journal of Zhejiang University (Engineering Science), 2021, 55(4): 684-694. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZDZC202104010.htm
[36]	陈涛, 郭卫卫, 孟令智, 等. 基于广义回归神经网络的路面摩擦系数预测模型[J]. 公路, 2014, 59(6): 1-6. https://www.cnki.com.cn/Article/CJFDTOTAL-GLGL201406001.htm CHEN Tao, GUO Weiwei, MENG Lingzhi, et al. Prediction model of road friction coefficient based on GRNN network[J]. Highway, 2014, 59(6): 1-6. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GLGL201406001.htm
[37]	SERIGOS P, ANDRE D, PROZZI J. Incorporating Surface microtexture in the prediction of skid resistance of flexible pavements[J]. Transportation Research Record, 2014, 2457: 105-113.
[38]	PATTANAIK M L, CHOUDHARY R, KUMAR B. Prediction of frictional characteristics of bituminous mixes using group method of data handling and multigene symbolic genetic programming[J]. Engineering With Computers, 2020, 36(4): 1875-1888.
[39]	LIU C Q, LI J, GAO J, et al. Three dimensional texture measurement using deep learning and multi-view pavement images[J]. Measurement, 2021, 172: 108828.
[40]	YANG G W, LI J Q, ZHAN Y, et al. Convolutional neural network-based friction model using pavement texture data[J]. Journal of Computing in Civil Engineering, 2018, 32(6): 04018052.
[41]	ZHAN Y, LI J Q, YANG G W, et al. Friction-ResNets: deep residual network architecture for pavement skid resistance evaluation[J]. Journal of Transportation Engineering, 2020, 146(3): 04020027.
[42]	张金喜, 郭旺达, 宋波, 等. 基于随机森林的沥青路面性能预测[J]. 北京工业大学学报, 2021, 47(11): 1256-1263. https://www.cnki.com.cn/Article/CJFDTOTAL-BJGD202111008.htm ZHANG Jinxi, GUO Wangda, SONG Bo, et al. Performance prediction of asphalt pavement based on random forest[J]. Journal of Beijing University of Technology, 2021, 47(11): 1256-1263. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-BJGD202111008.htm
[43]	ZHAN Y, LI J Q, LIU C, et al. Effect of aggregate properties on asphalt pavement friction based on random forest analysis[J]. Construction and Building Materials, 2021, 292: 123467.
[44]	YANG G W, YU W Y, LI J Q. Random forest-based pavement surface friction prediction using high-resolution 3D image data[J]. Journal of Testing and Evaluation, 2019, 49(2): 20180937.
[45]	高伟业, 张赛, 张杰, 等. 含湿相变粗糙多孔材质的热质耦合分形研究[J]. 应用数学和力学, 2022, 43(5): 561-568. doi: 10.21656/1000-0887.420328 GAO Weiye, ZHANG Sai, ZHANG Jie, et al. Thermo-mass coupling fractal study of wet phase-change rough porous materials[J]. Applied Mathematics and Mechanics, 2022, 43(5): 561-568. (in Chinese) doi: 10.21656/1000-0887.420328
[46]	张亚衡, 周宏伟, 谢和平. 粗糙表面分形维数估算的改进立方体覆盖法[J]. 岩石力学与工程学报, 2005, 24(17): 3192-3196. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX200517029.htm ZHANG Yaheng, ZHOU Hongwei, XIE Heping. Improved cubic covering method for fractal[J]. Chinese Journal of Rock Mechanics and Engineering, 2005, 24(17): 3192-3196. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX200517029.htm
[47]	郑少鹏, 程志豪, 陈亮亮, 等. 路面表面构造特征指标对抗滑性能的影响[J]. 公路, 2018, 63(6): 34-39. https://www.cnki.com.cn/Article/CJFDTOTAL-GLGL201806007.htm ZHENG Shaopeng, CHENG Zhihao, CHEN Liangliang, et al. Influence of pavement surface structural characteristic index on skid resistance[J]. Highway, 2018, 63(6): 34-39. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-GLGL201806007.htm
[48]	产品几何技术规范(GPS) 表面结构区域法: GB/T 33523.2—2017[S]. 2017. Geometrical product specifications (GPS) surface texture: areal: GB/T 33523.2—2017[S]. 2017. (in Chinese)
[49]	PARK J W, MOON Y S. Robust estimation of target scale by removing outlier motion vectors using MAD[J]. Electronics Letters, 2015, 51(9): 691-693.
[50]	BREIMAN L. Random forests[J]. Machine Learning, 2001, 45(1): 44-51.
[51]	杨彦从, 彭瑞东, 周宏伟. 三维空间数字图像的分形维数计算方法[J]. 中国矿业大学学报, 2009, 38(2): 251-258. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGKD200902019.htm YANG Yancong, PENG Ruidong, ZHOU Hongwei. Computation of fractal dimension for digital image in a 3-D space[J]. Journal of China University of Mining & Technology, 2019, 38(2): 251-258. (in Chinese) https://www.cnki.com.cn/Article/CJFDTOTAL-ZGKD200902019.htm