Foreign object debris (FOD) on airport runways is an important factor affecting aircraft flight safety, and current FOD detection technologies all have obvious deficiencies. In this paper, an indoor near-infrared (NIR) hyperspectral image data acquisition system with a wavelength range of 900-1700nm was built. The 14 samples of 6 common FODs and airport concrete runways were divided into reference and test sample sets, and the atlas data were collected for two common application scenarios. Preprocessing was performed on the reference sample set of hyperspectral images and reference spectral curves were extracted for 7 types of samples. Six spectral matching algorithms based on spectral angle matching (SAM), spectral information divergence (SID), spectral correlation coefficient (SCC) and their combinations are used to classify pixels one by one. By comparing the classification map, overall accuracy (OA), average accuracy (AA), and Kappa coefficient, a NIR hyperspectral FOD detection method based on SAM-SID (threshold Sc=40 pixel) criterion is obtained. The proposed method obtained ideal classification maps for the test sample set, with OA, AA and Kappa coefficients reaching 92%, 82% and 0.82, respectively, thus achieving good validation.
Damage to the skin coating of the Cessna 172R aircraft is an unavoidable and significant issue due to long flight training. Traditional detection methods are easy to missed and false detections, and hyperspectral technology can significantly improve. The spectral curves of the damaged and undamaged skin coating pixels in the near-infrared band (900-1700 nm) of the Cessna 172R aircraft skin samples were used to establish three spectral indices: ASCI-I, ASCI-II, and ASCI-III. Then the decision tree is used to perform recognition experiments on the two types of skin samples. The experimental results show that the decision tree model based on ASCI-I has the best recognition performance. Its global image recognition results are closely related to the spatial distribution of the actual targets. The Producer accuracy, User accuracy, and overall classification accuracy are all over 90%, and the Kappa coefficient is greater than 0.88.
KEYWORDS: Signal generators, Signal processing, Video, Heart, Interference (communication), Adversarial training, Gallium nitride, Education and training, Signal to noise ratio, Databases
Remote photoplethysmography (rPPG) is an optical technique that measures physiological signals from facial videos by analyzing subtle changes in the skin blood volume. However, rPPG signals generated in practical applications are easily affected by external environmental factors and the state of individuals, leading to irregular waveform variations that increase the difficulty in heart rate estimation. To improve the regularity of generated rPPG signals, we propose a standardized rPPG signal generation method. Specifically, facial videos are fed into the generator of a generative adversarial network (GAN) to predict a rough rPPG signal by supervised learning. In addition, a mathematical signal synthesizer model is used to generate noise-free standardized rPPG signals, which are subsequently fed into a discriminator along with the predicted signal for adversarial learning. This enables the generator to learn more standardized waveforms. As a result, the predicted signal waveform by the generator becomes closer to the waveform distribution of real rPPG signals. The proposed method is validated on the widely used MAHNOB-HCI, UBFC-rPPG, and MMSE-HR databases and shows significant improvements in the prediction accuracy and signal-to-noise ratio.
At present, there are few studies on nondestructive testing of aircraft surface based on hyperspectral imaging at home and abroad. Therefore, an indoor near infrared (NIR) hyperspectral damage detection system with a spectral resolution of 5nm was established, and the paint damage on the sample surface was identified. The reflectance calibration, average reflectance calculation and principal component analysis (PCA) dimensionality reduction were performed on the collected hyperspectral data. On this basis, the unsupervised classification iterative self-organizing Data analysis algorithm (ISODATA) is used to identify the damaged samples. The results show that the spectral curves of the damaged and undamaged pixels of the sample are significantly different at about 910nm. The first 10 principal components selected can contain 97% of the sample data information, which can realize the effective identification of damage samples based on ISODATA. In this study, paint damage was taken as an experimental sample to verify the feasibility of using near-infrared hyperspectral imaging technology for damage identification. In addition, preliminary outfield experiment results also show that it is feasible to apply this technology to aircraft surface damage detection.
Monitoring the formation process and occurrence state of methane in abyssal gas-liquid-hydrate coexistent system is the premise for gas hydrate research and exploitation, and the key lies in real time, synchronous and in-situ acquisition of multi state parameters, like concentration, temperature, pressure of methane. In this paper, we propose a novel multi parameter in situ methane sensor (Submarine Methane Imaging Interference Spectrometer, SMIIS) that can simultaneously measure concentration, temperature and pressure information of submarine methane. Then to evaluate SMIIS’s feasibility and performance, we build SMIIS’s simulation model and analyze its forward interferogram. The signal-to-noise ratios (SNRs) of the simulation interference fringes for the six spectral lines of methane are in the range of (3 - 618). The detection sensitivities for concentration, temperature and pressure measurements can reach to 0.5 nmol/L, 0.5 K, and 0.05 MPa, respectively. The results indicate that the preliminary design of SMIIS is feasible. After further testing and improvement, this system will have the potential to be applied to the seabed methane detection.
The continuous wavelet transform (CWT) introduces an expandable spatial and frequency window which can overcome the inferiority of localization characteristic in Fourier transform and windowed Fourier transform. The CWT method is widely applied in the non-stationary signal analysis field including optical 3D shape reconstruction with remarkable performance. In optical 3D surface measurement, the performance of CWT for optical fringe pattern phase reconstruction usually depends on the choice of wavelet function. A large kind of wavelet functions of CWT, such as Mexican Hat wavelet, Morlet wavelet, DOG wavelet, Gabor wavelet and so on, can be generated from Gauss wavelet function. However, so far, application of the Gauss wavelet transform (GWT) method (i.e. CWT with Gauss wavelet function) in optical profilometry is few reported. In this paper, the method using GWT for optical fringe pattern phase reconstruction is presented first and the comparisons between real and complex GWT methods are discussed in detail. The examples of numerical simulations are also given and analyzed. The results show that both the real GWT method along with a Hilbert transform and the complex GWT method can realize three-dimensional surface reconstruction; and the performance of reconstruction generally depends on the frequency domain appearance of Gauss wavelet functions. For the case of optical fringe pattern of large phase variation with position, the performance of real GWT is better than that of complex one due to complex Gauss series wavelets existing frequency sidelobes. Finally, the experiments are carried out and the experimental results agree well with our theoretical analysis.
AA2198 alloy is one of the third generation Al-Li alloys which have low density, high elastic modulus, high specific strength and specific stiffness. Compared With the previous two generation Al-Li alloys, the third generation alloys have much improved in alloys strength, corrosion resistance and weldable characteristic. For these advantages, the third generation Al-Li alloys are used as aircraft structures, such as C919 aviation airplane manufactured by China and Russia next generation aviation airplane--MS-21. As we know, the aircraft structures are usually subjected to more than 108 cycles fatigue life during 20-30 years of service, however, there is few reported paper about the third generation Al-Li alloys’ very high cycle fatigue(VHCF) which is more than 108 cycles fatigue. The VHCF experiment of AA2198 have been carried out. The two different initiation mechanisms of fatigue fracture have been found in VHCF. The cracks can initiate from the interior of the testing material with lower stress amplitude and more than 108 cycles fatigue life, or from the surface or subsurface of material which is the dominant reason of fatigue failures. During the experiment, the infrared technology is used to monitor the VHCF thermal effect. With the increase of the stress, the temperature of sample is also rising up, increasing about 15 °C for every 10Mpa. The theoretical thermal analysis is also carried out.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.