|
|
|
|
|
|
Feature Selection Algorithm for Identification of Male and Female
Cocoons Based on SVM Bootstrapping Re-Weighted Sampling |
CHEN Chu-han1, ZHONG Yang-sheng2, WANG Xian-yan3, ZHAO Yi-kun1, DAI Fen1* |
1. College of Electronic Engineering,South China Agricultural University,Guangzhou 510642, China
2. College of Animal Science,South China Agricultural University,Guangzhou 510642, China
3. Guangdong Sericulture Technology Promotion Center,Guangzhou 510640, China
|
|
|
Abstract The cost of identifying male and female cocoons by NIR is high,and the cost can be reduced by selecting useful features. Since there is a nonlinear relationship between the NIR spectra of female and male cocoons, a wrapper feature selection method, Bootstrapping Re-weighted Sampling Support Vector Machines (BRS-SVM), was proposed. The diffuse transmission NIR spectra of silkworm cocoons were collected by NirQuest512 NIR spectrometer. The heat map of characteristic importance was obtained by modeling the whole band of the test set, and the heat map obtained the range of important characteristic bands. Then, in the range of important characteristic bands, the single band features and continuous band area features were selected by BRS-SVM, Model-based ranking support vector machines (MBR-SVM), Model-based ranking Logistic Regression feature sorting method (MBR-LR), Recursive feature elimination (RFE), successive projections algorithm(SPA), Genetic Algorithm(GA), and then the support vector machines (SVM) and Logistic Regression (LR) sex classification models were established respectively. According to the characteristic importance heat map, it is found that the important area of male and female classification of silkworm cocoon was within 900~1 399 nm. We used this band to build the SVM model, and achieved 99.40% accuracy. BRS-SVM was used to select 5 single-band features. The accuracy of the test set is 89.56%, which is 2%~4% higher than other feature selection methods. RS-SVM was used to select 27 single-band features, and the accuracy of the test set of the SVM gender classification model was 94.97%, which reached the requirements of production conditions. The accuracy of modeling test set by BRS-SVM was 94.43% for 14 continuous band features. In the case of selecting a small number of features, our proposed BRS-SVM is superior to other methods. Using BRS-SVM to select a small number of features, we can establish a good performance of the female and male cocoon classification model, effectively reduce the cost, has important practical significance.
|
Received: 2021-07-13
Accepted: 2021-11-03
|
|
Corresponding Authors:
DAI Fen
E-mail: sunflower@scau.edu.cn
|
|
[1] FENG Wei-song(封槐松). China Sericultur(中国蚕业),2018,(1): 1.
[2] ZHAO Ming-yan,JIANG Xin-yu,NIU Bao-long,et al(赵明岩,蒋昕余,牛宝龙,等). Science of Sericulture(蚕业科学),2018,44(5): 711.
[3] Tao D,Wang Z,Li G,et al. Spectroscopy Letters,2018,51(8): 446.
[4] YAN Hui,LIANG Meng-xing,GUO Cheng,et al(颜 辉,梁梦醒,郭 成,等). Science of Sericulture(蚕业科学),2018,44(2): 283.
[5] Zhu Z,Yuan H,Song C,et al. Sensors and Actuators B: Chemical,2018,268: 299.
[6] DAI Fen,CHE Xin-xin,PENG Si-ran,et al(代 芬,车欣欣,彭斯冉,等). Journal of South China Agricultural University(华南农业大学学报),2018,39(2): 103.
[7] HONG Bin,DENG Bo,PENG Fu-yang,et al(洪 斌,邓 波,彭甫阳,等). Chinese Journal of Computer(计算机科学),2016,43(8): 19.
[8] LI Yu-qiang,PAN Tian-hong,LI Hao-ran,et al(李鱼强,潘天红,李浩然,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2019,39(12): 3809.
[9] JIAO Lei-zi,DONG Da-ming,ZHAO Xian-de,et al(矫雷子,董大明,赵贤德,等). Smart Agriculture(智慧农业),2020,2(2): 59.
[10] WANG Ling,LI Ding-ming,QIAN Hong-juan,et al(王 玲,李定明,钱红娟,等). Chinese Journal of Analysis Laboratory(分析试验室),2016,35(10): 1203.
[11] Shardlow M. An Analysis of Feature Selection Selection Techniques, Mathematics, Computer Sciences, 2011.
[12] Zhang J,Xiong Y,Min S. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy,2019,223: 117110.
|
[1] |
SHI Wen-qiang1, XU Xiu-ying1*, ZHANG Wei1, ZHANG Ping2, SUN Hai-tian1, 3, HU Jun1. Prediction Model of Soil Moisture Content in Northern Cold Region Based on Near-Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1704-1710. |
[2] |
WANG Xue-pei1, 2, ZHANG Lu-wei1, 2, BAI Xue-bing3, MO Xian-bin1, ZHANG Xiao-shuan1, 2*. Infrared Spectral Characterization of Ultraviolet Ozone Treatment on Substrate Surface for Flexible Electronics[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1867-1873. |
[3] |
WANG Yue1, 3, 4, CHEN Nan1, 2, 3, 4, WANG Bo-yu1, 5, LIU Tao1, 3, 4*, XIA Yang1, 2, 3, 4*. Fourier Transform Near-Infrared Spectral System Based on Laser-Driven Plasma Light Source[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1666-1673. |
[4] |
FENG Rui-jie1, CHEN Zheng-guang1, 2*, YI Shu-juan3. Identification of Corn Varieties Based on Bayesian Optimization SVM[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1698-1703. |
[5] |
YU Zhi-rong, HONG Ming-jian*. Near-Infrared Spectral Quantitative Analysis Network Based on Grouped Fully Connection[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1735-1740. |
[6] |
MENG Fan-jia1, LUO Shi1, WU Yue-feng1, SUN Hong1, LIU Fei2, LI Min-zan1*, HUANG Wei3, LI Mu3. Characteristic Extraction Method and Discriminant Model of Ear Rot of Maize Seed Base on NIR Spectra[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1716-1720. |
[7] |
PENG Yan-fang1, WANG Jun1, WU Zhi-sheng2*, LIU Xiao-na3, QIAO Yan-jiang2*. NIR Band Assignment of Tanshinone ⅡA and Cryptotanshinone by
2D-COS Technology and Model Application Tanshinone Extract[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(06): 1781-1785. |
[8] |
WANG Li-qi1, YAO Jing1, WANG Rui-ying1, CHEN Ying-shu1, LUO Shu-nian2, WANG Wei-ning2, ZHANG Yan-rong1*. Research on Detection of Soybean Meal Quality by NIR Based on
PLS-GRNN[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(05): 1433-1438. |
[9] |
FU Yan-hua1, LIU Jing2*, MAO Ya-chun2, CAO Wang2, HUANG Jia-qi2, ZHAO Zhan-guo3. Experimental Study on Quantitative Inversion Model of Heavy Metals in Soda Saline-Alkali Soil Based on RBF Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(05): 1595-1600. |
[10] |
ZHANG Zhao1, 2, 3, 4, YAO Zhi-feng1, 3, 4, WANG Peng1, 3, 4, SU Bao-feng1, 3, 4, LIU Bin3, 4, 5, SONG Huai-bo1, 3, 4, HE Dong-jian1, 3, 4*, XU Yan5, 6, 7, HU Jing-bo2. Early Detection of Plasmopara Viticola Infection in Grapevine Leaves Using Chlorophyll Fluorescence Imaging[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1028-1035. |
[11] |
LI Jia-yi1, YU Mei1, LI Mai-quan1, ZHENG Yu2*, LI Pao1, 3*. Nondestructive Identification of Different Chrysanthemum Varieties Based on Near-Infrared Spectroscopy and Pattern Recognition Methods[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1129-1133. |
[12] |
LI Xue-ying1, 2, LI Zong-min3*, CHEN Guang-yuan4, QIU Hui-min2, HOU Guang-li2, FAN Ping-ping2*. Prediction of Tidal Flat Sediment Moisture Content Based on Wavelet Transform[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1156-1161. |
[13] |
ZHANG Xiao-hong1, JIANG Xue-song1*, SHEN Fei2*, JIANG Hong-zhe1, ZHOU Hong-ping1, HE Xue-ming2, JIANG Dian-cheng1, ZHANG Yi3. Design of Portable Flour Quality Safety Detector Based on Diffuse
Transmission Near-Infrared Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(04): 1235-1242. |
[14] |
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. |
[15] |
CHENG Jie-hong1, CHEN Zheng-guang1, 2*, YI Shu-juan2. Wavelength Selection Algorithm Based on Minimum Correlation Coefficient for Multivariate Calibration[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2022, 42(03): 719-725. |
|
|
|
|