基于维纳估计的彩色图像光谱重建算法  

A Spectral Reflectance Reconstruction Algorithm for Color Image Using Wiener Estimation Method

作　　者：靖新[1] 岳田鑫 郑莉[1] 

机构地区：[1]沈阳建筑大学理学院,辽宁沈阳110168

出　　处：《沈阳建筑大学学报（自然科学版）》2014年第1期175-180,共6页Journal of Shenyang Jianzhu University：Natural Science

基　　金：国家自然科学基金项目(61272253);辽宁省自然科学基金项目(2013020013)

摘　　要：目的研究彩色图像光谱重建算法,解决彩色图像的RGB等色度参数在表征图像颜色信息时的局限性.方法采用维纳估计算法并结合多项式模型,对1 269个Munsell实验数据进行测试.结果光谱重建仿真实验中,对Munsell实验数据按照不同训练和检验样本来选取,当数码相机RGB信号扩展项数为10时,均方根误差为0.024 4,优于扩展项为3和7时的精度,可见扩展项数对光谱重建精度有直接影响.结论结合多项式模型的维纳估计算法在进行彩色图像光谱反射比重建的研究中,扩展项的增加可以有效地改善光谱重建效果.A spectral reflectance reconstruction algorithm for color image has been studied in this paper, in order to solve the problem that the chrominance parameters RGB cannot give color image information fully and accurately. 1 269 Munsell experimental data have been tested using Wiener estimation algorithm com- bined with the polynomial model. According to different training and testing samples in spectral reconstruc- tion experiments, the corresponding Munsell experimental data are selected. When the extensive item number of the RGB signal in digital camera is 10, simulation illustrates that the root mean square error is 0. 0244, better than the results of extensive items 3 and 7. It is shown that the spectral reconstruction accuracy can be improved with the high items. The results suggest that, based on the new algorithm using Wiener estimation method and polynomial model, the spectral reconstruction accuracy can be improved efficiently with the large number of items.

关 键 词：彩色图像 维纳估计 多项式模型 光谱反射比 光谱重建 

分 类 号：TP334.2[自动化与计算机技术—计算机系统结构]