img

QQ群聊

img

官方微信

高级检索

黄金科学技术 ›› 2023, Vol. 31 ›› Issue (1): 153-162.doi: 10.11872/j.issn.1005-2518.2023.01.148

• 采选技术与矿山管理 • 上一篇    

基于FCM-WA联合算法的多种类矿石图像分割

汤文聪1(),罗小燕1,2()   

  1. 1.江西理工大学机电工程学院,江西 赣州 341000
    2.江西省矿冶机电工程研究中心,江西 赣州 341000
  • 收稿日期:2022-10-17 修回日期:2022-11-28 出版日期:2023-02-28 发布日期:2023-03-27
  • 通讯作者: 罗小燕 E-mail:1976678028@qq.com;lxy9416@163.com
  • 作者简介:汤文聪(1997-),男,广东韶关人,硕士研究生,从事图像处理研究工作。1976678028@qq.com
  • 基金资助:
    国家自然科学基金项目“基于多尺度内聚颗粒模型的振动破碎能耗研究”(51464017);江西省教育厅科学技术项目“黑钨磨矿过程状态监测与负荷智能识别”(200827)

Image Segmentation of Multi Kinds of Ores Based on FCM-WA Joint Algorithm

Wencong TANG1(),Xiaoyan LUO1,2()   

  1. 1.College of Mechanical and Electrical Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, Jiangxi, China
    2.Jiangxi Mining and Metallurgy Engineering Research Center, Ganzhou 341000, Jiangxi, China
  • Received:2022-10-17 Revised:2022-11-28 Online:2023-02-28 Published:2023-03-27
  • Contact: Xiaoyan LUO E-mail:1976678028@qq.com;lxy9416@163.com

摘要:

矿石图像分割是基于机器视觉的矿石粒度分布检测的重要组成部分。针对复合矿山中颜色多样、纹理复杂且边缘粘连的多种类矿石图像难以识别与分割的问题,提出了一种基于FCM-WA联合算法的矿石图像分割方法。首先对矿石图像进行形态学优化,利用双边滤波、直方图均衡化和形态学重构来优化矿石图像的几何特征,减少噪声对分割效果的影响,提高图像对比度;然后将模糊C均值聚类(FCM)算法与分水岭(WA)算法相结合,利用FCM算法进行聚类迭代,计算出合适的分割阈值并对矿石图像进行分割,输出二值化图像;再利用基于距离变换的WA算法优化FCM算法的分割结果,对FCM算法输出的矿石图像边缘粘连部分进行分割,以获取最佳的分割图像。研究结果表明:(1)利用形态学优化流程处理矿石图像能够减少噪声并增强边缘信息,从而提高对比度;(2)相比传统的大津法和遗传算法,本文所提FCM-WA方法的稳健性更强、分割效果更好,对多种类的矿石图像像素分割准确率和矿石粒度识别准确率均可达到92%以上;(3)通过试验验证,FCM-WA方法能够精确地分割颜色多样、纹理特征复杂及边缘粘连的多种类矿石图像,分割结果满足粒度分布检测的要求;(4)FCM-WA方法符合现实矿山企业生产的需求,能够为研发新型矿山智能化粒度检测设备提供可靠的技术支持。

关键词: 复合矿山, 矿石图像, 形态学处理, 模糊C均值聚类, 分水岭算法, 边缘分割

Abstract:

Ore image segmentation is an important part of ore size distribution detection based on machine vision.In order to solve the problem that it is difficult to recognize and segment the multi kinds of ore images with various colors,complex textures and adhesive edges in composite mines,a method of ore image segmentation based on FCM-WA combined algorithm was proposed.Firstly,the ore image is optimized by morphology,which uses bilateral filtering,histogram equalization and morphological reconstruction to optimize the geometric features of the ore image,reduce the impact of noise on the segmentation effect,and improve the image contrast.Then,the FCM algorithm was combined with the watershed algorithm,and the FCM algorithm was used for clustering iteration to calculate the appropriate segmentation threshold,segment the ore image,and output the binary image.Then,the WA algorithm based on distance transformation was used to optimize the segmentation result of FCM algorithm,and the edge conglutination part of ore image output by FCM algorithm was segmented to obtain the best segmentation image.The results show that:(1)Using morphological optimization process to process ore images can reduce noise,enhance edge information and improve contrast.(2) Compared with the traditional Otsu method and genetic algorithm,the FCM-WA method in this paper is more robust and has better segmentation effect.The accuracy of pixel segmentation and ore particle size recognition for multiple kinds of ore images can reach more than 92%.(3) The experiment results show that the FCM-WA method can accurately segment many kinds of ore images with diverse colors,edge adhesion and complex texture features,and the segmentation results meet the requirements of particle size distribution detection.(4) The FCM-WA method in this paper is in line with the production needs of real mining enterprises,and can provide reliable technical support for the development of new mine intelligent particle size detection equipment.

Key words: compound mines, ore image, morphological treatment, fast and robust fuzzy c-means clustering algorithm, watershed algorithm, edge segmentation

中图分类号: 

  • TP391

图1

多种类矿石图像"

图2

形态学优化处理流程"

图3

图像采集系统"

图4

形态学优化结果"

图5

基于FCM-WA联合算法的多阈值分割流程"

图6

各分割流程效果"

图7

FCM-WA算法收敛效果"

图8

不同算法分割效果"

表1

不同算法分割指标对比"

算法

像素数

/个

正确分割像素数/个

TPR

/%

矿石个数正确识别个数

TOR

/%

OTSU2 320 4701 930 96183.212689635.8
GA2 320 4702 224 51895.8626813048.5
FCM2 320 4702 245 42496.7626810739.9
FCM-WA2 320 4702 198 47194.7426826297.7

表2

4种方法的性能指标对比"

图像像素分割准确率(TPR矿石粒度识别准确率(TOR
OTSUGAFCMFCM-WAOTSUGAFCMFCM-WA
183.295.896.794.735.848.539.997.7
289.897.398.696.238.445.538.492.4
384.995.097.194.534.337.733.496.1
488.496.497.795.139.248.244.894.3
591.497.798.996.544.345.242.195.9
Cai Gaipin, Liu Zhan, Wang Long,et al,2020.Segmentation of watershed ore image with marker based on morphological optimization [J].Science,Technology and Engineering,20(23):9497-9502.
Chen Zhikun, Jiang Junjun, Jiang Xinwei,et al,2020.A robust hyperspectral remote sensing image feature extraction method based on improved bilateral filtering [J].Journal of Wuhan University (Information Science Edition),45(4):504-510.
Deng Wenjing, Zhou Wu, Cai Xiaoshu,2019.Multi dimensional feature KFCM clustering segmentation algorithm for color image of core particles [J].China Powder Technology,25(6):12-18.
Huang H, Meng F, Zhou S,et al,2019.Brain image segmentation based on FCM clustering algorithm and rough set[J].IEEE Access,7:12386-12396.DOI:10.1109/ACCESS.2019.2893063 .
doi: 10.1109/ACCESS.2019.2893063
Huang M L, Liu Y L, Yang Y M,2022.Edge detection of ore and rock on the surface of explosion pile based on improved canny operator[J].Alexandria Engineering Journal,61(12):10769-10777..
Li Guoyao, Wang Teng,2020.Research on concrete crack detection based on morphological treatment and feature analysis [J].Building Structure,50(Supp.2):529-533.
Li H X, Wang X L, Yang C H,et al,2021.Ore image segmentation method based on GAN-UNet[J].Control Theory and Application,38(9):1393-1398.DOI:10.7641/CTA.2021.00558 .
doi: 10.7641/CTA.2021.00558
Li H, Pan C, Chen Z,et al,2020.Ore image segmentation method based on U-Net and watershed[J].Computers,Materials and Continua,65(1):563-578.DOI:10.32604/cmc.2020.09806 .
doi: 10.32604/cmc.2020.09806
Lin Y F, Diao Y, Du Y Z,et al,2021.Automatic cell counting for phase-contrast microscopic images based on a combination of Otsu and watershed segmentation method[J].Microscopy Research and Technique,85(1):169-180. .
Lin Y F, Fang C F, Gao L Z,2022.Adhesive abrasive detection for diamond images based on improved watershed algorithm[C]//Journal of Physics:Conference Series.IOP Publishing,2289(1):012023.DOI:10.1088/1742-6596/2289/1/012023 .
doi: 10.1088/1742-6596/2289/1/012023
Liu X B, Zhang Y, Jing H,et al,2020.Ore image segmentation method using U-Net and Res_Unet convolutional networks[J].RSC Advances,10(16):9396-9406.DOI:10.1039/c9ra05877j .
doi: 10.1039/c9ra05877j
Kanhui Lü, Zhang Daxing,2021.Infrared image enhancement algorithm based on adaptive histogram equalization coupled with Laplace transform [J].Optical Technology,47(6):747-753.
Qin G F, Li Q T,et al,2019.Pavement image segmentation based on fast FCM clustering with spatial information in internet of things[J].Multimed Tools and Applications,78(5):5181-5191.DOI:10.1007/s11042-017-4683-0 .
doi: 10.1007/s11042-017-4683-0
Raju P, Rao V M, Rao B P,2019.Optimal GLCM combined FCM segmentation algorithm for detection of kidney cysts and tumor[J].Multimedia Tools and Applications,78(13):18419-18441..
Ruan Qiuqi, Ruan Yuzhi,2020.Digital Image Processing (Fourth Edition) (US) Gonzalez [M].Beijing:Electronic Industry Press.
Verma H, Verma D, Tiwari P K,2021.A population based hybrid FCM-PSO algorithm for clustering analysis and segmentation of brain image[J].Expert Systems with Applications,167:114121..
Wang W, Li Q, Xiao C,et al,2021.An improved boundary-aware U-Net for ore image semantic segmentation[J].Sensors,21(8):2615..
Wang Wei, Li Qing, Zhang Dezheng,et al,2023.A survey of ore image processing based on deep learning[J].Chinese Journal of Engineering,45(4):621-631.
Xiao D, Liu X, Le B T,et al,2020.An ore image segmentation method based on RDU-Net model[J].Sensors,20(17):4979..
Zhan Y, Zhang G,2019.An improved OTSU algorithm using hi-stogram accumulation moment for ore segmentation[J].Sym-metry,11(3):431..
Zhang G, Li M, Zhan Y,et al,2017.Ore image thresholding segmentation using double windows with fifisher discrimination[C]//The 13th International Conference on Natural Computation,Fuzzy Systems and Knowledge Discovery(ICNC-FSKD).New York:IEEE: 2715-2719.DOI:10.1109/ FSKD.2017.8393208 .
doi: 10.1109/ FSKD.2017.8393208
Zhang Jianli, Feng Xiaoyu, Zhang Jianqiang,2022.Application of lifting wavelet and watershed algorithm in ore particle sized detection[J].Machinery Designed and Manufacture,(6):290-294.
Zhang Jianli, Sun Shenshen, Qin Shuqi,2019.Ore image segmentation based on optimal threshold segmentation of genetic algorithm [J].Science,Technology and Engineering,19(7):105-109.
Zhou J, Yang M,2022.Bone region segmentation in medical images based on improved watershed algorithm[J].Computational Intelligence and Neuroscience. .
蔡改贫,刘占,汪龙,等,2020.基于形态学优化处理的标记符分水岭矿石图像分割[J].科学技术与工程,20(23):9497-9502.
陈志坤,江俊君,姜鑫维,等,2020.一种基于改进双边滤波的鲁棒高光谱遥感图像特征提取方法[J].武汉大学学报(信息科学版),45(4):504-510.
邓文晶,周骛,蔡小舒,2019.岩心颗粒彩色图像的多维特征KFCM聚类分割算法[J].中国粉体技术,25(6):12-18.
李国耀,王腾,2020.基于形态学处理与特征分析的混凝土裂缝检测研究[J].建筑结构,50(增2):529-533.
吕侃徽,张大兴,2021.基于自适应直方图均衡化耦合拉普拉斯变换的红外图像增强算法[J].光学技术,47(6):747-753.
阮秋琦,阮宇智,2020.数字图像处理(第四版)(美)冈萨雷斯[M].北京:电子工业出版社.
王伟,李擎,张德权,等,2023.基于深度学习的矿石图像处理研究综述[J].工程科学学报,45(4):621-631.
张建立,冯小丽,张建强,2022.提升小波和分水岭算法在矿石粒度检测中的应用[J].机械设计与制造,(6):290-294.
张建立,孙深深,秦书棋,2019.基于遗传算法最佳阈值分割的矿石图像分割[J].科学技术与工程,19(7):105-109.
[1] 聂振宇,周科平,梁志鹏. 基于VR技术的矿山冒顶片帮事故教学培训[J]. 黄金科学技术, 2021, 29(4): 620-628.
[2] 郑高华,王雨琦,王毓华,卢东方,郑霞裕. 基于虚拟现实技术的磨矿分级工艺自主设计系统的开发[J]. 黄金科学技术, 2021, 29(1): 120-128.
[3] 马宁,胡乃联,李国清,郭对明,侯杰. 基于模糊层次分析法的高原矿井人机功效评价[J]. 黄金科学技术, 2019, 27(6): 871-878.
[4] 荆永滨, 赵新涛, 冯兴隆. 节理岩体矿岩块度三维模拟研究[J]. 黄金科学技术, 2018, 26(3): 357-364.
[5] 朱忠华,王李管,陶干强,蒲成志. 自然崩落法一体化放矿优化控制与智能化管理系统研制[J]. 黄金科学技术, 2017, 25(6): 83-91.
[6] 张二洋,陈建宏. 基于Surpac矿山设计软件及虚幻引擎实现的矿山虚拟现实漫游系统[J]. 黄金科学技术, 2017, 25(4): 93-98.
[7] 谭正华,谭姣月,潘梅,王李管. 基于空间轮廓线的高质量矿体表面三维重构方法[J]. 黄金科学技术, 2017, 25(2): 96-103.
[8] 胡笑坤,宋慧昌,刘青. 金矿选矿厂磨矿分级过程仿真系统研发[J]. 黄金科学技术, 2016, 24(5): 94-101.
[9] 韩可琦, 胡吉锋, 刘华东, 刘银志, 梅文泽, 曹其俭. 煤矿采掘计划计算机仿真系统实现[J]. J4, 1999, 7(4~5): 55-58.
[10] 黄勇. 信息技术在露天采矿业中的应用[J]. J4, 1999, 7(4~5): 74-77.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!