基于铸体薄片图像颜色空间与形态学梯度的岩石分类

来源期刊:中南大学学报(自然科学版)2016年第7期

论文作者:刘烨 程国建 马微 郭超

文章页码:2375 - 2383

关键词:岩石图像;分类;颜色空间;形态学梯度;支持向量机

Key words:rock images; classification; color space; morphological gradient; SVM

摘    要:为实现自动高效且结果可靠的岩石分类,提出一种基于岩石薄片图像的自动分类方法。该方法通过偏光显微镜采集的铸体薄片图像,由图像的原始颜色空间与其形态学梯度中提取特征参数,统计各图像矩阵分量的标准算术值构建岩石分类的特征空间,利用支持向量机方法建立特征空间与岩石类别之间的映射关系。采用鄂尔多斯盆地苏里格地区的100幅沉积岩岩石薄片图像对该方法进行测试。研究结果表明:该自动分类方法结果的正确率达95%以上,说明结合颜色空间与形态学梯度的高效岩石图像自动分类方法具有较高的准确性与可靠性。

Abstract: A new high efficiency automatic rock classification method for thin section images of rocks was proposed to solve the problems caused by subjective errors and the difference of capture equipments and conditions. This classification method, whose basic data were from polarizing microscope, was based on original color space and morphology gradient features to build the relationship between these features extracted from images and rock types with SVM. Practical test data set composed of 100 image samples was from Sulige gas field in Ordos basin. The results show that the accuracy of this method reaches 95%, which proves the method is stable and dependable both in theoretical and practical aspects.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号