| 混合Gamma分布的LJ1-01夜间灯光去噪算法——以沪宁杭都市圈为例 |
| |
| 引用本文： | 杨鹏,刘靖钰,刘德儿,邹纪伟,张荷苑.混合Gamma分布的LJ1-01夜间灯光去噪算法——以沪宁杭都市圈为例[J].测绘科学,2021,46(2):113-121. |
| |
| 作者姓名： | 杨鹏 刘靖钰 刘德儿 邹纪伟 张荷苑 |
| |
| 作者单位： | 江西理工大学土木与测绘工程学院,江西赣州 341000;成都大学,成都610106 |
| |
| 基金项目： | 国家自然科学基金项目(41361077,41561085);江西省自然科学基金项目(20161BAB203091)。 |
| |
| 摘 要： | 为了将人类活动引起的有效光点与非人类活动引起的噪声点区分开来、提高特征目标提取和分析中的提取精度,该文提出了一种混合卡方分布算法对LJ1-01夜光数据进行去噪。由于LJ1-01数据符合卡方分布,但单卡方拟合精度较低,因此使用多个卡方的叠加来表示LJ1-01数据和噪声子集,并基于最小二乘原理,结合有效统计灯光值构造间接平差方程,求解各叠加态的比例;然后,基于比例构建具有高拟合精度的连续概率密度曲线;最后,构建了丰度函数,并以原始图像中不同噪声子集的权重中心的偏移量去除噪声。为了验证算法的通用性和稳定性,选择不同的噪声子集并与正态分布算法对比,实验结果表明该文算法对噪声去除效果较好。
|
| 关 键 词： | 夜间灯光 LJ1-01 混合卡方分布 去噪 间接平差 |
LJ1-01 nightlight denoising algorithm based on mixed Gamma distribution——taking Shanghai-Nanjing-Hangzhou metropolitan area as an example |
| |
| YANG Peng,LIU Jingyu,LIU Deer,ZOU Jiwei,ZHANG Heyuan.LJ1-01 nightlight denoising algorithm based on mixed Gamma distribution——taking Shanghai-Nanjing-Hangzhou metropolitan area as an example[J].Science of Surveying and Mapping,2021,46(2):113-121. |
| |
| Authors: | YANG Peng LIU Jingyu LIU Deer ZOU Jiwei ZHANG Heyuan |
| |
| Institution: | (School of Civil and Surveying&Mapping Engineering,Jiangxi University of Science and Technology,Ganzhou,Jiangxi 341000,China;Chengdu University,Chengdu 610106,China) |
| |
| Abstract: | In order to distinguish effective spots caused by human activities from noise points caused by non-human activities and improve the extraction accuracy of feature target extraction and analysis,this paper proposes a hybrid Chi-square distribution algorithm to de-noise LJ1-01 noctilucent data.Since the LJ1-01 data conforms to the Chi-square distribution,but the single Chi-square fitting accuracy is low,the superposition of multiple Chi-squares is used to represent the LJ1-01 data and noise subset.Based on the least square principle,the indirect adjustment equation is constructed by combining with the effective statistical light value to solve the proportion of each superposition state.Then,the continuous probability density curve with high fitting accuracy is constructed based on proportion.Finally,the abundance function is constructed and the noise is removed by the offset of the weight center of different noise subsets in the original image.In order to verify the universality and stability of the algorithm,different subsets of noise were selected and compared with the normal distribution algorithm,experimental results showed that the algorithm in this paper had a better effect on noise removal. |
| |
| Keywords: | nightlight LJ1-01 hybrid Chi-square distribution denoising indirect adjustment |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
