|
|
|
|
|
|
Research on Spectral Reflectance Reconstruction Sample Selection Based on NSGA-Ⅱ Algorithm |
LIN Lu, WANG Zhi-feng*, LI Chao |
School of Computer and Software Engineering, University of Science and Technology Liaoning, Anshan 114051, China
|
|
|
Abstract To solve the problems of heavy workload and low reconstruction accuracy caused by the large number of training samples in the process of spectral reflectance reconstruction, a sample selection method for spectral reflectance reconstruction based on the NSGA-Ⅱ algorithm was proposed. Recently, Liang et al. proposed a new method to select representative samples from many sample data. The spectral reconstruction error was defined as the product of the root mean square error and goodness-of-fit coefficient. Based on this standard, the sample with the smallest spectral reconstruction error was selected as the representative sample for spectral reflectance reconstruction. Inspired by the work of Liang et al., this method combined with the NSGA-Ⅱ algorithm to select representative samples. First, the polynomial regression algorithm and pseudo-inverse method were used to reconstruct spectral reflectance for all training samples. Then, the NSGA-Ⅱ algorithm was used to set two objective functions. One is the sum of the spectral root mean square error of the required number of representative samples; the other is the sum of the reciprocal of the goodness-of-fit coefficients, which minimizes the values of the two objective functions. All samples with Pareto level 1 selected by NSGA-Ⅱ algorithm are sorted according to the occurrence times of samples from high to low and selected as representative samples from high to low until the required number of representative samples is reached. Suppose the number of representative samples selected from the sample set with Pareto level 1 is less than the required number. In that case,the samples that appear most frequently in the next level and have not been selected are selected until the number of representative samples reaches the demand. The experiment divided 1 269 dull Munsell standard color cards into even color cards and odd color cards according to sample subscript. In the first group of experiments, Munsell odd color cards were used as the whole training samples, and 20 color blocks were randomly selected from Munsell even color cards as the test samples. In the second group of experiments, Munsell even color cards were selected as the whole training samples, and 20 color blocks were randomly selected from Munsell odd color cards as the test samples. The third group of experimental training samples is the same as the first group, and the RC24 color card is the test sample. The proposed method is compared with the three sample selection methods proposed by Mohammadi, Cao and Liang. The experimental results show that the NSGA-Ⅱ algorithm combined with polynomial regression and pseudo-inverse method to select representative samples for spectral reflectance reconstruction is superior to the existing sample selection methods in terms of root mean square error and color difference, and this method is not for a specific system has generality.
|
Received: 2022-10-13
Accepted: 2023-08-05
|
|
Corresponding Authors:
WANG Zhi-feng
E-mail: wangzhifeng_sia@126.com
|
|
[1] WANG Wen-ju, WANG Jiang-wei(王文举, 王江薇). Packaging Engineering(包装工程), 2020, 41(11): 254
[2] Mohammadi M, Nezamabadi M, Berns R S, et al. Spectral Imaging Target Development Based on Hierarchical Cluster Analysis, 12th Color and Imaging Conference, 2004: 59.
[3] Shen Huiliang, Zhang Honggang, Xin John H, et al. Applied Optics, 2008, 47(13): 494.
[4] Eckhard T, Valero E M, Hernádez-Andrés J, et al. Applied Optics,2014, 53(4): 709.
[5] Cao Bin, Liao Ningfang, Cheng Haobo. Color Research & Application, 2017, 42(3): 327.
[6] Liang Jinxing, Zhu Qiang, Liu Qiang, et al. Color Research & Application, 2021, 47(1): 107.
[7] Zhang Lu, Wang Chengqun, Fang Mengyuan, et al. IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, 2022, 105(8): 1175.
[8] LI Fu-hao, LI Chang-jun(李富豪, 李长军). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2021, 41(10): 3281.
[9] GiagkiozisI, Fleming P J. Information Sciences, 2015, 293: 338.
[10] LI Sui-xian(李遂贤). Acta Optica Sinica(光学学报), 2020, 40(4): 62.
|
[1] |
MA Yuan, LI Ri-hao, ZHANG Wei-feng*. Research on the Training Samples Selection for Spectral Reflectance
Reconstruction Based on Improved Weighted Euclidean Distance[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(12): 3924-3929. |
[2] |
JIANG Wan-li1, 2, SHI Jun-sheng1, 2*, JI Ming-jiang1, 2. Establishment of Visible and NIR Spectral Reflectance Database of Plant Leaves and Principal Component Analysis[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(08): 2366-2373. |
[3] |
JIANG Dan-yang1, WANG Zhi-feng1*, GAO Cheng1, 2, LI Chang-jun1. Spectral Reflectance Reconstruction With Color Constancy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1044-1048. |
[4] |
ZHENG Kai-yi1, ZHANG Wen1, DING Fu-yuan1, ZHOU Chen-guang1, SHI Ji-yong1, Yoshinori Marunaka2, ZOU Xiao-bo1*. Using Ensemble Refinement (ER) Method to OptimizeTransfer Set of Near-Infrared Spectra[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1323-1328. |
[5] |
ZHAO Xiao-kang, ZHAO Xin, ZHU Qi-bing*, HUANG Min. A Model Construction Method of Spectral Nondestructive Detection for Apple Quality Based on Unsupervised Active Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(01): 282-291. |
[6] |
LI Fu-hao, LI Chang-jun*. Spectral Reflectance Reconstruction Based on Camera Raw RGB Using Weighted Third-Order Polynomial and Wiener Estimation[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(10): 3281-3285. |
[7] |
ZHENG Kai-yi1, FENG Tao1, ZHANG Wen1, HUANG Xiao-wei1, LI Zhi-hua1, ZHANG Di1, SHI Ji-yong1, Yoshinori Marunaka2, ZOU Xiao-bo1*. Weighted SPXYE (WSPXYE) and Its Application to Transfer Set Selection in Near Infrared Spectra[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2021, 41(03): 984-989. |
[8] |
LU De-jun, CUAN Kai-xuan, ZHANG Wei-feng*. Research on Spectral Reflectance Estimation Using Locally Weighted Linear Regression within k-Nearest Neighbors[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2018, 38(12): 3708-3712. |
[9] |
JIN Zhao-xi1, ZHANG Xiu-juan2, LUO Fu-yi2, AN Dong1, 3*, ZHAO Sheng-yi1, RAN Hang1, YAN Yan-lu1 . Study of Modeling Samples Selection Method Based on Near Infrared Spectrum [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(12): 3920-3925. |
[10] |
LI Chan,WAN Xiao-xia*,LIU Qiang*, LIANG Jin-xing, LI Jun-feng. Research on the Training Samples Selection for Spectral Reflectance Reconstruction Based on Principal Component Analysis[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(05): 1400-1405. |
[11] |
YANG Ping1, LIAO Ning-fang1, SONG Hong2 . Study of Approaches to Spectral Reflectance Reconstruction Based on Digital Camera [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2009, 29(05): 1176-1180. |
|
|
|
|