基于NVST观测的米粒识别和形态特征分析
发布时间：2018-07-16 08:40
【摘要】：米粒组织是太阳表层对流运动所产生的一种形态特征.由于米粒强度分布不均匀以及边缘比较模糊,使得采用传统的基于强度和梯度阈值的方法来准确地识别它们变得困难.因此,本文提出了一个基于相位一致性的米粒识别算法.选用中国科学院云南天文台抚仙湖的新一代太阳真空望远镜(New Vaccum Sloar Telescope,简称NVST)的高分辨活动区观测资料来展示算法的识别过程,并根据识别结果分析了米粒的形态特征.选取了2个目前已存在的米粒识别算法来验证所提算法的准确性和有效性,实验结果表明所提算法能够有效地提取那些低对比度的米粒特征.同时为了检验算法对阈值的响应程度,分别选取了3组阈值来检验不同阈值情况下的差异性.结果证明所提算法对阈值响应不敏感.为了进一步检验所提算法获取统计结果的准确性,对米粒的直径、强度、形状以及分形维数进行统计.统计结果很好地符合了已有文献的结论,这也进一步验证所提算法的准确性和健壮性,能够用于进一步的科学研究.
[Abstract]:Rice grain tissue is a morphological feature produced by the convective motion of the surface layer of the sun. Because of the uneven distribution of grain strength and fuzzy edges, it is difficult to identify them accurately by using traditional methods based on intensity and gradient threshold. Therefore, this paper proposes a phase consistency based rice grain recognition algorithm. The high-resolution active region observation data of the New Generation of New Vaccum Sloar Telescope( NVST) at Fuxian Lake, Yunnan Observatory, Chinese Academy of Sciences, are used to demonstrate the recognition process of the algorithm. The morphological characteristics of rice grains are analyzed according to the recognition results. Two existing rice grain recognition algorithms are selected to verify the accuracy and validity of the proposed algorithm. The experimental results show that the proposed algorithm can extract the low-contrast rice grain features effectively. At the same time, in order to test the response of the algorithm to the threshold, three groups of thresholds were selected to test the differences under different threshold conditions. The results show that the proposed algorithm is insensitive to threshold response. In order to check the accuracy of the proposed algorithm, the diameter, intensity, shape and fractal dimension of the grain were analyzed. The statistical results are in good agreement with the existing conclusions, which further verify the accuracy and robustness of the proposed algorithm, and can be used for further scientific research.
【作者单位】： 昆明理工大学信息工程与自动化学院 云南省计算机技术应用重点实验室;
【基金】：国家自然科学基金(U1231205, U1531132, 11463003, 11573012, 11303011)资助
【分类号】：P182.2
本文编号：2125858
[Abstract]:Rice grain tissue is a morphological feature produced by the convective motion of the surface layer of the sun. Because of the uneven distribution of grain strength and fuzzy edges, it is difficult to identify them accurately by using traditional methods based on intensity and gradient threshold. Therefore, this paper proposes a phase consistency based rice grain recognition algorithm. The high-resolution active region observation data of the New Generation of New Vaccum Sloar Telescope( NVST) at Fuxian Lake, Yunnan Observatory, Chinese Academy of Sciences, are used to demonstrate the recognition process of the algorithm. The morphological characteristics of rice grains are analyzed according to the recognition results. Two existing rice grain recognition algorithms are selected to verify the accuracy and validity of the proposed algorithm. The experimental results show that the proposed algorithm can extract the low-contrast rice grain features effectively. At the same time, in order to test the response of the algorithm to the threshold, three groups of thresholds were selected to test the differences under different threshold conditions. The results show that the proposed algorithm is insensitive to threshold response. In order to check the accuracy of the proposed algorithm, the diameter, intensity, shape and fractal dimension of the grain were analyzed. The statistical results are in good agreement with the existing conclusions, which further verify the accuracy and robustness of the proposed algorithm, and can be used for further scientific research.
【作者单位】： 昆明理工大学信息工程与自动化学院 云南省计算机技术应用重点实验室;
【基金】：国家自然科学基金(U1231205, U1531132, 11463003, 11573012, 11303011)资助
【分类号】：P182.2
【相似文献】
相关硕士学位论文 前1条
1 陈宇超;NVST多通道高分辨观测系统软件设计和实现[D];中国科学院研究生院(云南天文台);2015年
，本文编号：2125858
本文链接：https://www.wllwen.com/kejilunwen/tianwen/2125858.html
