当前位置 首页 >写作素材 >

基于协同表示的多特征融合岩石分类

作者:jkyxc 浏览数:


打开文本图片集

摘要:针对传统的岩石薄片成分分析耗时、识别率不高等问题,提出了一种基于协同表示(CR)的岩石薄片成分分析方法。首先,分析探讨了岩石薄片中颗粒纹理特性,证明将薄片图像的分层多尺度局部二值化(HMLBP)特征与灰度共生矩阵(GLCM)特征相融合能有效地表征岩石薄片中颗粒的纹理。然后,为降低识别阶段时间复杂度,采用主成分分析(PCA)方法将新特征降维到100维。最后,采用基于协同表示的分类器(CRC)进行分类识别。与基于稀疏表示的分类器(SRC)分别采用样本字典中某一个样本单独编码表征预测样本不同,基于协同表示的分类器采用样本字典中的所有样本协同编码表征预测样本,借助不同样本的同一属性提高识别率。实验结果表明该方法的识别速度较基于稀疏的分类器识别方法提高300%,识别率提高2%;在实践应用中能较好地区分岩石薄片中的石英成分和长石成分。

关键词:协同表示;纹理特征;特征融合;分类器;岩石薄片

中图分类号: TP391.41 文献标志码:A

0引言

岩石粒度分析与成分识别是沉积学研究的一个重要方面。国内外学者对于岩石颗粒成分分析开展了大量的研究工作[1-3]。Kachanuban等[4]通过主成分分析和改进的空间频率测量方法对天然岩石纹理图像进行识别,该方法将有限种岩石纹理图像粗略的分成三大类:均匀的岩石图像、非均匀深色岩石或含少量黑色颗粒的明亮岩石图像、含有许多细小或者大的黑色颗粒非均匀的明亮岩石图像。Shang等[5]通过对岩石纹理特征分析,利用支持向量机(Support Vector Machine, SVM)分类器对14种岩石纹理图像进行分类,取得了较好效果。Izadi等[6]根据石英和正长石在单偏光和正交偏光下表现出的不同特性,利用灰度共生矩阵(Gray Level Cooccurrence Matrix, GLCM)中的逆差矩对二者进行分类,同时利用人工神经网络对三角石和辉石进行了识别,取得了较好效果。Li等[7]通过正交偏光下碎屑岩薄片中岩石颗粒干涉色的变化特性,利用地理信息系统对其进行分割识别。Aprile等[8]采用神经网络对岩石中的石英和钙聚合物进行识别分类,并取得了较好效果。

上述方法从宏观层面对岩石成分进行分析并取得了一定效果,应用于微观分析时仍具有一定局限性,效果并不理想。因此本文基于实验室自主研发的岩石成分微观分析系统,提出了一种基于协同表示的识别方法(Collaborative Representation based Classification, CRC),从微观层面对岩石

成分进行研究。该系统能拍摄不同光照角度下岩石薄片的偏光序列图,且序列图间位移形变较小。特征提取采用分层多尺度局部二值化(Hierarchical Multiscale Local Binary Pattern, HMLBP)和灰度共生矩阵(GLCM)相融合的方法。经过大量实验证明,该方法具有较高识的别率和较强的鲁棒性,时间代价也大幅度减小。

4结语

一直以来岩石颗粒识别都是地质学上的一个重要研究课题,传统的人工识别耗时费力,效率不高。通过图像处理的方法对颗粒进行区分是现行较好的研究方法。众多学者的共同研究取得了许多显著成果。以图像处理为基础实现岩石分类的智能化是发展的最终方向。

本文通过对岩石特征提取及分类方法的研究,提出了一种新的岩石颗粒识别方法。结合实验室自主研发的微观图像分析系统,通过对不同角度偏光图像识别的结果进行统计,最终得出分类结果。实验证明该方法取得了较好的识别效果,解决了岩石薄片训练样本不足的问题,有助于石油地质行业从业人员分析地质薄片中的岩石组成成分。

本文提出的方法取得了较好的效果,但是仍然存在有不足,如何在特征提取阶段提取更准确完整的信息并实现参数设置智能化,使得提取效果最佳是以后的研究方向。

参考文献:

[1]

GORSEVSKI P V, ONASCH C M, FARVER J R, et al. Detecting grain boundaries in deformed rocks using a cellular automata approach [J]. Computers and Geosciences, 2012,42(3):136-142.

[2]

ZHOU Y, STARKEY J, MANSINHA L. Segmentation of petrographic images by integrating edge detection and region growing [J]. Computers and Geosciences, 2004,30(8):817-831.

[3]

HEILBRONNER R. Automatic grain boundary detection and grain size analysis using polarization micrographs or orientation images [J]. Journal of Structural Geology, 2000,22(7):969-981.

[4]

KACHANUBAN T, UDOMHUNSAKUL S. Natural rock images classification using spatial frequency measurement [C]// Proceedings of the 2007 International Conference on Intelligent and Advanced Systems. Piscataway, NJ: IEEE, 2007:815-818.

[5]

SHANG C, BARNES D. Support vector machinebased classification of rock texture images aided by efficient feature selection [J]. World Congress Computational Intelligence. Brisbane, QLD:2012:1-8.

SHANG C, BARNES D. Support vector machinebased classification of rock texture images aided by efficient feature selection [C]// Proceedings of the 2012 International Joint Conference on Neural Networks. Piscataway, NJ: IEEE, 2012:1-8.

[6]

IZADI H, SADRI J, MEHRAN N A. A new approach to apply texture features in minerals identification in petrographic thin sections using ANNs [C]// Proceedings of the 2013 8th Iranian Conference on Machine Vision and Image Processing. Piscataway, NJ: IEEE, 2013:257-261.

[7]

LI B, ZHANG T, BAI L, et al. Components classification of the clastic rock thinsections based on GIS [C]// Proceedings of the 2010 International Conference on Image Analysis and Signal Processing. Piscataway, NJ: IEEE, 2010:327-331.

[8]

APRILE A, CASTELLANO G, ERAMO G. Combining image analysis and modular neural networks for classification of mineral inclusions and pores in archaeological potsherds [J]. Journal of Archaeological Science, 2014,50:262-272.

[9]

ULABY F T, KOUYATE F, BRISCO B, et al. Textural information in SAR images [J]. IEEE Transactions on Geoscience and Remote Sensing, 1986,24(2):235-245.

[10]

OJALA T, PIETIKINEN M, MENP T. Multiresolution grayscale and rotation invariant texture classification with local binary patterns [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,24(7):971-987.

[11]

GUO Z, ZHANG L, ZHANG D, et al. Hierarchical multiscale LBP for face and palmprint recognition [C]// Proceedings of the 17th IEEE International Conference on, Tsinghua University Department of Computing, the Hong Kong Polytechnic University Institute of Image Processing and Recognition, Xi’an Jiaotong University Image Processing (ICIP) :2010, 26-29:1522-4880

GUO Z, ZHANG L, ZHANG D, et al. Hierarchical multiscale LBP for face and palmprint recognition [C]// Proceedings of the 2010 17th IEEE International Conference on Image Processing. Piscataway, NJ: IEEE, 2010:4521-4524.

[12]

WRIGHT J, YANG A Y, GANESH A, et al. Robust face recognition via sparse representation [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009:31(2):210-227.

[13]

ZHANG L, YANG M, FENG X. Sparse representation or collaborative representation: which helps face recognition? [C]// Proceedings of the 2011 International Conference on Computer Vision. Washington, DC: IEEE Computer Society, 2011:471-478.

background

This work is supported by the National Natural Science Foundation of China (61372174).

LIU Juexian, born in 1987, M. S. candidate. His research interests include image processing and pattern recognition.

TENG Qizhi, born in 1962, Ph. D., professor. Her research interests include image processing, pattern recognition, threedimensional reconstruction.

WANG Zhengyong, born in 1969, Ph. D., associate professor. Her research interests include image processing and pattern recognition.

HE Xiaohai, born in 1964, Ph. D., professor. His research interests include image processing, pattern recognition and network communication.

推荐访问:岩石 协同 融合 特征 分类

相关文章:

Top