|
|
|
|
|
|
| Near-Infrared Random Forest Classification and Recognition Based on Multi-Feature Fusion |
| XIE Xi-ru1, LUO Hai-jun1, 2*, LI Guo-nan1, FAN Xin-yan1, WANG Kang-yu1, LI Zhong-hong1, WANG Jie1 |
1. Chongqing Key Laboratory of Optoelectronic Functional Materials，School of Physics and Electronic Engineering，Chongqing Normal University，Chongqing 401331，China
2. Chongqing National Center for Applied Mathematics，Chongqing Normal University，Chongqing 401331，China
|
|
|
|
|
Abstract To solve the problem of low data classification accuracy and poor model stability of functional near-infrared brain functional imaging signals in brain-computer interfaces, this paper proposes to study the changes of cerebral blood oxygen concentration in the prefrontal brain region during the stimulation period and carry out a study of the state binary classification identification of task and rest on the experimental data of the prefrontal brain region, and take the extracted single feature and multi-feature fusion as the inputs of the model, respectively, and validate the results of the classification through the model The conjecture that multi-feature fusion can improve the classification accuracy to some extent. Firstly, this paper designs an experimental paradigm for the prefrontal brain region: word fluency cognition experiment. The device used to collect the data is NirSmart, a Danyang Huichuang device from China, with a sampling rate of 11 Hz and a spatial resolution of 3 cm. The device uses avalanche diode and ultramicro photo detection technology, and the sensitivity is up to 0.1 pW. After that, the collected data are preprocessed using the Homer2 toolbox, the features are extracted using MATLAB, and the Random Forest model is constructed using two metrics of feature importance and error curve to evaluate the classification results. Importance and error curves were used to evaluate the performance of the model. Finally, the average of 20 runs of the random forest model was used as the final classification result. The experimental results show that multiple features can improve the final classification result compared with a single feature. The best second classification result for the case of three-feature fusion is 93.84%, 2.32%, 4.25%, and 5.33%, higher than the single-feature mean, slope, and peak-to-peak value, respectively. From the results of the experimental data, it can be seen that multi-feature fusion can improve the accuracy of near-infrared brain functional imaging classification to a certain extent. The performance of the random forest model is stable, which is expected to solve the previous problems of low classification accuracy and poor model stability and promote the development and application of brain-computer interface systems.
|
|
Received: 2024-01-15
Accepted: 2024-04-25
|
|
|
|
Corresponding Authors:
LUO Hai-jun
E-mail: luohaijun@cqnu.edu.cn
|
|
[1] Hamaoka Takafumi，McCully Kevin K. The Journal of Physiological Sciences，2019，69(6): 799.
[2] HONG Zi-yun，YAN Chen-lin，MIN Hong，et al(洪子云，严承琳，闵 红，等). Spectroscopy and Spectral Analysis（光谱学与光谱分析），2022，42(9): 2800.
[3] LIU Guang-da，CAO Ze-yuan，ZHOU Ge，et al(刘光达，曹泽元，周 葛，等). Optical Precision Engineering（光学精密工程），2016，24（8）：1846.
[4] Coyle Shirley，Ward Tomás，Markham Charles，et al. Physiological Measurement，2004，25(4): 815.
[5] Power Sarah D，Falk Tiago H，Chau Tom. Journal of Neural Engineering，2010，7(2): 026002.
[6] Schudlo Larissa C，Chau Tom. Journal of Neural Engineering，2014，11(1): 016003.
[7] Erdoğan S B，Yükselen G. Sensors，2022，22(14): 5407.
[8] Wang Feng，Ding Haishu，Tian Fenghua，et al. Physiological Measurement，2001，22(1): 201.
[9] Sitaram Ranganatha，Zhang Haihong，Guan Cuntai，et al. Neurolmage, 2007，34(4): 1416.
[10] CHU Bao，HUANG Yao，NI Jing-shu，et al(储 宝，黄 尧，倪敬书，等). Chinese Journal of Lasers（中国激光），2022，49(5): 0507303.
[11] DING Heng-bing，YE Fei-yue（丁恒兵，叶飞跃）. Computer Engineering and Design（计算机工程与设计），2023，44(11): 3373.
[12] ZHANG Yue-mei，LIU Yuan-hua（张月梅，刘媛华）. Computer & Digital Engineering (计算机与数字工程)，2020，48(2): 367.
|
| [1] |
MAO Li-yu1, 2, BIN Bin1*, ZHANG Hong-ming2*, LÜ Bo2, 3*, GONG Xue-yu1, YIN Xiang-hui1, SHEN Yong-cai4, FU Jia2, WANG Fu-di2, HU Kui5, SUN Bo2, FAN Yu2, ZENG Chao2, JI Hua-jian2, 3, LIN Zi-chao2, 3. Development of Wheat Component Detector Based on Near Infrared
Spectrum[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2768-2777. |
| [2] |
JIANG Xiao-gang1, 2, HE Cong1, 2, JIANG Nan3, LI Li-sha1, ZHU Ming-wang1, LIU Yan-de1, 2*. Discrimination of Apple Origin and Prediction of SSC Based on
Multi-Model Decision Fusion[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2812-2818. |
| [3] |
ZHANG Ting-lin, TANG Long, PENG Dong-yu, TANG Hao, JIANG Pan-pan, LIU Bo-tong, CHEN Chuan-jie*. A Diagnostic Method for Electron Density of Plasmas by
Machine-Learning Combined With Stark Broadening[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2778-2784. |
| [4] |
ZHOU Lei-jinyu1, ZHOU Li-na1*, CHEN Li-mei1, KONG Li-juan1, QIAO Jian-lei2, LI Ming-tang3. Visible/Near Infrared Spectroscopic Modeling for Cadmium
Contaminated Lettuce[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2805-2811. |
| [5] |
MU Liang-yin1, ZHAO Zhong-gai1*, JIN Sai2, SUN Fu-xin2, LIU Fei1. Near-Infrared Prediction Models for Quality Parameters of Culture Broth in Seed Tank During Citric Acid Fermentation[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2819-2826. |
| [6] |
JIANG Yu-heng1, YAN Bo1, ZHUANG Qing-yuan1, WANG Ai-ping1, CAO Shuang1, TIAN An-hong1, 2, FU Cheng-biao1*. Quantitative Inversion Model of Soil Heavy Metals Zn and Ni Based on Fractional Order Derivative[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2850-2857. |
| [7] |
GUAN Cheng1, LIU Ming-yue1, 2, 3, 4*, MAN Wei-dong1, 2, 3, 4, ZHANG Yong-bin1, ZHANG Qing-wen1, FANG Hua1, LI Xiang1, GAO Hui-feng1. Estimation of Chlorophyll Content in Spartina Alterniflora Leaves Based on Continous Wavelet Transformation and Random Forest Algorithm[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2993-3000. |
| [8] |
LIU Yu-juan1, 2, 3, LIU Yan-da1, 2, 3, YAN Zhen1, 4, ZHANG Zhi-yong1, 2, 3, CAO Yi-ming1, 2, 3, SONG Ying1, 2, 3*. Classification of Hybrid Convolution Hyperspectral Images Based on
Attention Mechanism[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2916-2922. |
| [9] |
GUO Zhi-qiang1, ZHANG Bo-tao1, ZENG Yun-liu2*. Study on Sugar Content Detection of Kiwifruit Using Near-Infrared
Spectroscopy Combined With Stacking Ensemble Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2932-2940. |
| [10] |
XIE Yu-yu1, 2, 3, CHEN Zhi-hui2, HOU Xue-ling1, 3, LIU Yong-qiang1, 3*. A New Method for Determination of Eburicoic Acid in Fomes Officinalis Ames by NIR Combined With PLS[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2981-2987. |
| [11] |
LI Xuan1, GAN Shu1, 2*, YUAN Xi-ping2, 3, 4, YANG Min3, 4, GONG Wei-zhen1. Spectral Characteristic and Identification Modelling of Three Typical Wetland Vegetation Along the Seashore of the East Coast of the Erhai Lake[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(09): 2439-2444. |
| [12] |
ZHU Yu-kang1, LU Chang-hua1, ZHANG Yu-jun2, JIANG Wei-wei1*. Quantitative Method to Near-Infrared Spectroscopy With Multi-Feature Fusion Convolutional Neural Network Based on Wavelength Attention[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(09): 2607-2612. |
| [13] |
MAO Ya-chun1, WEN Jie1*, CAO Wang1, DING Rui-bo1, WANG Shi-jia2, FU Yan-hua3, XU Meng-yuan1. Fusion Algorithm Research Based on Imaging Spectrum of Anshan Iron Ore[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(09): 2620-2625. |
| [14] |
WENG Ding-kang1, FAN Zheng-xin1, KONG Ling-fei1, SUN Tong1*, YU Wei-wu2. Rapid Identification of Shelled Bad Torreya Grandis Seeds Based on
Visible-Near Infrared Spectroscopy and Chemometrics[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(09): 2675-2682. |
| [15] |
WU Bin1, XIE Chen-ao2, CHEN Yong2, WU Xiao-hong2, JIA Hong-wen1. Discrimination of Chuzhou Chrysanthemum Tea Grades Using Noise
Discriminant C-Means Clustering[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(08): 2202-2207. |
|
|
|
|
