Rock particle size analysis is an important work in rock slice image analysis.In order to solve the problem of poor segmentation effect of fine rock particles in rock slice image by traditional threshold method,a rock slice image segmentation scheme based on deep learning is proposed.Firstly,U-Net is used to extract the cast area in the image.Then,the RCF network is used to obtain the edge of rock particles.Finally,the particles edge and cast area information are used to segment the particles.The experimental results show that compared with the threshold segmentation method,this method performs better in the segmentation of fine rock particles,and has higher segmentation accuracy at the edge of rock particles.
关键词
岩石铸体薄片图像/U-Net/RCF/岩石颗粒分割
Key words
rock cast slice image/U-Net/RCF/rock particle segmentation
维普
万方数据