橡胶复合材料中炭黑微观结构图像的拟合算法  

Fitting Algorithm of Microstructure Image of Carbon Black Reinforced Rubber Composites

作　　者：何红[1,2] 陈增云 张亚茹 章易慎 张立群[1] 李凡珠[1] HE Hong;CHEN Zengyun;ZHANG Yaru;ZHANG Yishen;ZHANG Liqun;LI Fanzhu(Beijing University of Chemical Technology,Beijing 100029,China)

机构地区：[1]北京化工大学有机无机复合材料国家重点实验室,北京100029 [2]北京化工大学机电工程学院,北京100029

出　　处：《橡胶工业》2023年第1期68-74,共7页China Rubber Industry

基　　金：国家重点研发计划项目(2018YFB1502501)。

摘　　要：将炭黑聚集体视为由多个圆形原生粒子构成,对橡胶复合材料中炭黑聚集体形态进行图像拟合分析。基于炭黑补强橡胶复合材料的微观结构图像,在采用图像分割和阈值迭代等方法处理图像背景缺陷的基础上,研究了轮廓骨架算法、最大内切圆算法和K-means聚类算法3种拟合算法处理炭黑聚集体图像,并用峰值信噪比和结构相似度2个指标对图像拟合效果进行评价。结果表明,轮廓骨架算法拟合炭黑聚集体形态效果最优,更适用于炭黑补强橡胶复合材料微观结构重构时对炭黑聚集体形态的描述。Carbon black aggregates were considered to be composed of multiple circular primary particles,and the morphology of carbon black aggregates in rubber composites was analyzed by image fitting.Based on the microstructure image of carbon black reinforced rubber composites,three fitting algorithms,contour skeleton algorithm,maximum inscribed circle algorithm and K-means clustering algorithm to deal with carbon black aggregate morphology were studied based on using image segmentation and threshold iteration and other methods to handle image background defects.The image fitting effects were evaluated by two indicators of peak signal-to-noise ratio(PSNR)and structural similarity(SSIM).The results showed that the contour skeleton algorithm had the best effect in fitting the morphology of carbon black aggregates,and it was more suitable for describing the morphology of carbon black aggregates in the microstructure reconstruction of carbon black reinforced rubber composites.

关 键 词：橡胶复合材料 炭黑补强 炭黑聚集体 微观结构 图像处理 拟合算法 轮廓骨架算法 峰值信噪比 结构相似度 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]