摘要
提出了一种基于K-Means的颜色量化算法的岩石图像预处理方法,利用K-Means自动将图像中相似的颜色合并为一种颜色,减少了岩石图像中的无用颜色。对于颜色量化,假设RGB岩石图像中的每个像素点都有自己的颜色,并且每个像素点都有对应坐标,使用欧几里得距离公式计算每个像素点之间的距离,最后利用K-Means算法对这些有特定颜色的坐标点进行无监督聚类,从而实现颜色的量化。应用结果表明,将基于K-Means的颜色量化算法用在岩石图像中,可以减少岩石图像中的颜色数量,且能在较低性能的计算机设备中很好地再现岩石图像,同时也能提高岩石图像处理的效率。
A rock image preprocessing method based on K-Means color quantization algorithm is proposed.The similar colors in the image are merged into one color automatically by K-Means to reduce the useless colors in the rock image.For color quantization,assuming that each pixel in RGB rock image has itself color and corresponding coordinates,the distance between pixels can be calculated using Euclidean distance formula.Finally,the unsupervised cluster of these coordinate points with specific color is finished using K-Means algorithm so as to achieve color quantization.The application results show that to apply the color quantization algorithm based on K-Means in rock image processing can reduce the number of colors in rock image,make rock image reproduced well in low-performance computer equipment,and improve the efficiency of rock image processing.
作者
程国建
魏珺洁
CHENG Guojian;WEI Junjie(College of Computer,Xi'an Shiyou University,Xi'an 710065,Shaanxi,China)
出处
《西安石油大学学报（自然科学版）》
CAS
北大核心
2019年第3期114-119,共6页
Journal of Xi’an Shiyou University（Natural Science Edition）
基金
国家科技重大专项(2011ZX05044)
陕西省工业科技攻关项目(2015GY104)
