|
|
|
|
|
|
| Near Infrared Hyperspectral Identification of Surface Damage on Aircraft Wings |
| LIU Qing-song1, DU Wen-jing1, LUO Bo2, LI Kai-ge1, DAN You-quan1*, XU Luo-peng1, YANG Xiu-feng2, TANG Shen-lan1 |
1. School of Science & Key Laboratory of Photonic and Optical Detection in Civil Aviation, Civil Aviation Flight University of China, Guanghan 618307, China
2. Maintenance Department, Civil Aviation Flight University of China, Guanghan 618307, China
|
|
|
|
|
Abstract The development of detection techniques and equipment for aircraft surface damage has significant practical significance for flight safety and operational efficiency. Spectral matching technology is a crucial technology that must be addressed in the hyperspectral detection of aircraft surface damage. The recognition accuracy of different spectral matching algorithms often varies depending on the research object. To use a spectral matching algorithm to achieve damage identification of aircraft samples, this article first built an indoor near-infrared hyperspectral system for aircraft surface damage detection and collected hyperspectral data of reference samples and skin samples. It produced standard spectra of two types of pixels using damaged pixel spectra and non-destructive pixel spectra. Subsequently, based on the matching method for calculating the similarity between the measured pixel spectrum and the standard spectrum, four types of single spectrum matching algorithms, namely spectral angle (SA), Mahalanobis distance (MD), spectral information divergence (SID), and spectral correlation coefficient (SCC), and six types of combined spectrum matching algorithms, were used for damage identification of two types of aircraft samples. The accuracy of multiple spectral matching algorithms' damage identification results was evaluated using the overall classification accuracy Pa and Kappa coefficient. A reasonable threshold group that can better meet the detection requirements is provided by optimizing the threshold parameters of single algorithms, such as SA, MD, SID, and SCC. Furthermore, based on the above four types of single matching algorithms, six types of combined matching algorithms were designed and used for sample damage identification, like SA-MD, SA-SID, SA-SCC, MD-SID, MD-SCC, and SID-SCC. The results show that the identification accuracy of those combined algorithms is relatively higher than that of any matching algorithms. Finally, this article presents the optimal single matching algorithm and combination matching algorithm for damage identification of aircraft samples, with the SCC algorithm and MD-SCC algorithm achieving damage identification rates of over 95% and 97.5% for both types of samples, respectively. This can provide technical support for hyperspectral detection of aircraft surface damage in the outfield.
|
|
Received: 2023-06-06
Accepted: 2024-01-15
|
|
|
|
Corresponding Authors:
DAN You-quan
E-mail: dandin12x@163.com
|
|
[1] PAN Qiao, YANG Li-yuan, SUN Zhi-hua，et al(潘 峤, 杨丽媛, 孙志华, 等). Environmental Testing(环境试验), 2020, (2): 41.
[2] Zhang Tianyu, Zhang Teng, He Yuting, et al. Chinese Journal of Aeronautics, 2023, 36(4): 1.
[3] WU Jun, LI Xin, LIU Shao-yu, et al(吴 军, 李 鑫, 刘少禹, 等). Acta Optica Sinica(光学学报), 2021, 41(11): 1115002.
[4] Jovancevic I, Larnier S, Orteu J J, et al. Journal of Electronic Imaging, 2015, 24(6): 061110.
[5] CHEN Yun-quan, LI Shuang-gao, HUANG Xiang，et al(陈允全, 李泷杲, 黄 翔, 等). Aeronautical Manufacturing Technology(航空制造技术), 2019, 62(10): 51.
[6] BAI Xue-bing, MA Dian-kun, ZHANG Meng-jie, et al(白雪冰, 马殿坤, 张梦杰, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2022, 42(7): 1993.
[7] Ma Yandy, Mannion Anthony, O'Brien Stephen. Proc. of SPIE, 2018, 10819: 108190X-1.
[8] GAO Jian-shu, ZHENG Da-chuan, YU Zhi-jing, et al(高建树, 郑大川, 于之靖, 等). Laser & Infrared(激光与红外), 2014, 44(4): 391.
[9] SU Hong-jun, DU Pei-jun(苏红军, 杜培军). Remote Sensing Technology and Application(遥感技术与应用), 2006, 21(4): 288.
[10] YANG Guo-peng, YU Xu-chu, LIU Wei, et al(杨国鹏, 余旭初, 刘 伟, 等). Bulletin of Surveying and Mapping(测绘通报), 2007, (10): 17.
[11] WEI Xiang-po, YU Xu-chu, FU Qiong-ying，et al(魏祥坡, 余旭初, 付琼莹, 等). Geography and Geo-Information Science(地理与地理信息科学), 2016, 32(3): 29.
[12] XU Long-xin, SUN Yong-hua, HE Shi-jun，et al(徐隆鑫, 孙永华, 何仕俊, 等). Journal of Capital Normal University(Natural Science Edition)[首都师范大学学报(自然科学版)], 2021, 42(6): 50.
[13] Du Wenjing, Liu Qingsong, Li Kaige, et al. Proc. of SPIE, 2023, 12558: 125580J-1.
|
| [1] |
WANG Cai-ling, ZHANG Guo-hao. IPSO-BPNN: A Quantitative Model for Nitrite Content in Water Quality Using Transmissive Spectroscopy Combined With Improved Particle Swarm Optimization and Backpropagation Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3172-3178. |
| [2] |
LI Zhi-yuan1, TIAN An-hong1, 2*. Quantitative Prediction and Spatial Distribution of Soil Heavy Metal Zn Based on Spectral Indices[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3287-3293. |
| [3] |
PENG Bo1, WEN Zhao-yang1, WEN Qi1, LIU Ting-ting1, 2*, XING Shuai3, WU Teng-fei3, YAN Ming1, 2*. Research of Mid-Infrared Time-Stretch Frequency Upconversion
Hyperspectral Imaging System[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3037-3042. |
| [4] |
WENG Shi-zhuang, PAN Mei-jing, TAN Yu-jian, ZHANG Qiao-qiao, ZHENG Ling*. Prediction of Soluble Solid Content in Apple Using Image Spectral Super-Resolution[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3095-3100. |
| [5] |
LI Bin1, 2, LU Ying-jun2, SU Cheng-tao2, LIU Yan-de1, 2. Detection of Different Levels of Damage in Gong Pears Based on
Reflectance/Absorbance/Kubelka-Munk Spectroscopy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3101-3108. |
| [6] |
SHI Rui1, 2, ZHANG Han2, WANG Cheng1, 2, KANG Kai2, LUO Bin1, 2*. Detection of Wheat Single Seed Vigor Using Hyperspectral Imaging and Spectrum Fusion Strategy[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3206-3212. |
| [7] |
DENG Zhi-gang1, 2, ZHAO Hong-mei2*, ZHA Wen-xian2, TANG Lin-ling2, TIAN Ye2. A Hyperspectral Vegetation Feature Band Selection Based on Quantum
Genetic Spectral Angle Mapper Algorithm[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3258-3265. |
| [8] |
ZHONG Qing1, Mamattursun EZIZ1, 2*, Mireguli AINIWAER1, 2, HOU Mao-rui3, LI Hao-ran4. Hyperspectral Inversion of Cobalt Content in Urban Soils[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3266-3272. |
| [9] |
GUO Hong-xu1, WANG Long1, YANG Kai1, WU Fan1, DENG Yi-rong2, TANG Chang-cheng1, CHEN Zhi-liang3*, XIAO Rong-bo1*. Research on Combination Optimization of Hyperspectral Inversion Model for Soil Cr Contamination[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3273-3279. |
| [10] |
LIU Ying1, 2, LIU Yu1, YUE Hui1, 2*, BI Yin-li2, 3, PENG Su-ping4, JIA Yu-hao1. Remote Sensing Inversion of Soil Carbon Emissions in Desertification Mining Areas[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2840-2849. |
| [11] |
YUE Ji-bo1, LENG Meng-die1, TIAN Qing-jiu2, GUO Wei1, LIU Yang3, FENG Hai-kuan4, QIAO Hong-bo1*. Estimation of Leaf Physical and Chemical Parameters Based on Hyperspectral Remote Sensing and Deep Learning Technologies[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2873-2883. |
| [12] |
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. |
| [13] |
DENG Yun1, 2, WU Wei1, 2, SHI Yuan-yuan3, CHEN Shou-xue1, 2*. Red Soil Organic Matter Content Prediction Model Based on Dilated
Convolutional Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2941-2952. |
| [14] |
FAN Jie-jie1, 2, QIU Chun-xia1, FAN Yi-guang2, CHEN Ri-qiang2, LIU Yang2, BIAN Ming-bo2, MA Yan-peng2, YANG Fu-qin4, FENG Hai-kuan2, 3*. Wheat Yield Prediction Based on Continuous Wavelet Transform and
Machine Learning[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(10): 2890-2899. |
| [15] |
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. |
|
|
|
|
