不同偏光系统下的岩石薄片图像联合分割

7Combined segmentation of rock slice images under different polarizing systems

针对传统切换单偏光与正交偏光系统进行镜下岩石薄片分析和含量估计耗时耗力,且准确率较低的问题,在色度亮度空间设计一种岩石薄片图像联合分割方法:先将相同视域下的单偏光图像与正交偏光图像在色度亮度空间中分别进行分解,然后对色度图像进行双边滤波处理,抑制噪声和去除纹理,计算滤波后两个色度图像之间的差值,并依次使用K-Means聚类算法和分水岭分割算法得到最终分割结果。结果表明,使用单偏光岩石薄片图像与正交偏光岩石薄片图像联合分割较大提升了矿物颗粒计数准确率和分割结果的交并比。研究结果可为地质工作者鉴定岩石薄片提供指导。

To solve the problem of low accuracy and time-consuming in the traditional switching between single polarization and orthogonal polarization systems for the analysis of rock slices and content estimation under microscope,a combined segmentation method of rock slice image in chrominance luminance space was designed.Firstly,the single polarized image and the orthogonal polarized image from the same field of view were decomposed separately in the chromatic and bright space.Then,the chromaticity image was sub‐jected to bilateral filtering to suppress noise and remove texture.The difference between the filtered two chromaticity images was calculated,and the final segmentation result was obtained by using K-Means clus‐tering algorithm and watershed segmentation algorithm in sequence.The research results can provide guid‐ance for geologists to identify rock slices.