...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Laser Physics: An International Journal devoted to Theoretical and Experimental Laser Research and Application >Study on the effects of sample selection on spectral reflectance reconstruction based on the algorithm of compressive sensing
【24h】

Study on the effects of sample selection on spectral reflectance reconstruction based on the algorithm of compressive sensing

机译：基于压缩感知算法的样本选择对光谱反射率重建的影响研究

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In order to solve the problem that reconstruction efficiency and precision is not high, in this paper different samples are selected to reconstruct spectral reflectance, and a new kind of spectral reflectance reconstruction method based on the algorithm of compressive sensing is provided. Four different color numbers of matte color cards such as the ColorChecker Color Rendition Chart and Color Checker SG, the copperplate paper spot color card of Panton, and the Munsell colors card are chosen as training samples, the spectral image is reconstructed respectively by the algorithm of compressive sensing and pseudo-inverse and Wiener, and the results are compared. These methods of spectral reconstruction are evaluated by root mean square error and color difference accuracy. The experiments show that the cumulative contribution rate and color difference of the Munsell colors card are better than those of the other three numbers of color cards in the same conditions of reconstruction, and the accuracy of the spectral reconstruction will be affected by the training sample of different numbers of color cards. The key technology of reconstruction means that the uniformity and representation of the training sample selection has important significance upon reconstruction. In this paper, the influence of the sample selection on the spectral image reconstruction is studied. The precision of the spectral reconstruction based on the algorithm of compressive sensing is higher than that of the traditional algorithm of spectral reconstruction. By the MATLAB simulation results, it can be seen that the spectral reconstruction precision and efficiency are affected by the different color numbers of the training sample.
机译：为了解决重建效率和精度不高的问题，本文选择了不同的样本来重建光谱反射率，并提出了一种基于压缩感知算法的新型光谱反射率重建方法。选择哑光色卡（ColorChecker Color Rendition Chart和Color Checker SG），潘通（Panton）铜版纸专色卡和芒塞尔（Munsell）色卡这四种不同色号作为训练样本，分别通过以下算法重建光谱图像：压缩感测和伪逆和维纳，并将结果进行比较。通过均方根误差和色差精度评估这些光谱重建方法。实验表明，在相同的重建条件下，孟塞尔色卡的累积贡献率和色差均优于其他三张色卡，并且光谱重建的准确性将受到训练样本的影响。不同数量的色卡。重建的关键技术意味着训练样本选择的统一性和代表性对重建具有重要意义。本文研究了样本选择对光谱图像重建的影响。基于压缩感知算法的频谱重建精度高于传统的频谱重建算法。从MATLAB仿真结果可以看出，光谱重建的精度和效率受训练样本不同色号的影响。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号