This PDF file contains the front matter associated with SPIE Proceedings Volume 13494, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

This study proposes a new strategy to accurately classify γ-TiAl samples with different microstructures using laser-induced breakdown spectroscopy (LIBS) combined with deep learning techniques. We first observed the microstructure of six groups of γ-TiAl treated with different solid solution temperatures and found that the percentage of lamellae increased with increasing temperature, while the percentage of γ phase substantially decreased. Next, the elemental characteristic spectral lines were collected by a coaxial acquisition device. Then we performed baseline correction and normalization on the LIBS spectra to eliminate the background signals. Principal Component Analysis (PCA) was then used to reduce the dimensionality to simplify the data structure. Finally, the processed data were fed into three different deep learning models, namely, Multilayer Perceptron (MLP), Long Short-Term Memory Network (LSTM), and Convolutional Neural Network (CNN), for training and classification. The classification accuracy using MLP, LSTM, and CNN was 83.33%, 81.87%, and 80.42%, respectively. The effect of material microstructure characterization by LIBS spectroscopy combined with the PCA-MLP model is particularly remarkable. This study provides a new solution for the rapid analysis of microstructures of engineering materials.

Ultraviolet differential optical absorption spectroscopy (UV-DOAS) has been successfully applied in long-distance detection of polluting gases in atmosphere. However, the traditional DOAS method will fail if gas absorption signals are annihilated in noisy signals. To address this issue, chlorine (Cl₂) was selected as the target gas. In this paper, a DOAS experimental system for open-path detection was developed, and an algorithm for decoupling slow-changing signals of Cl₂ and particle matters (PM) was proposed. Firstly, the weight analysis of full spectrum band was carried out based on synergy interval partial least squares (Si-PLS). As a result, low SNR band, Cl₂ feature absorption band and non-feature band were matched according to their contribution. Secondly, an equivalent scattering model was established to obtain the light intensity attenuation of PM in the non-feature band. Finally, the signal decoupling was realized by deducting the PM attenuation spectrum from the absorption band of Cl₂. This study provides a new feasible method for large scale detection of Cl₂ leakage in open environments.

Due to the growing societal demand for public safety and environmental protection, there is an increasing focus on the technology of rapid on-site bacterial detection. The present study aims to establish an analytical method using microscopic confocal Raman spectroscopy combined with chemometrics to accurately identify different bacteria, thus enabling nondestructive testing. This paper examines the Raman spectra of Gram-negative bacteria Escherichia coli, Gram-positive bacteria Staphylococcus aureus, Bacillus cereus, and others. The spectra were obtained using a microscopic confocal Raman spectrometer and analyzed for their characteristics. The Matlab-based PCA method reduces the dimensionality of the data. Six pattern recognition algorithms in chemometrics, namely linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), k-nearest neighbor (kNN), naive Bayesian (NB) model, classification decision tree (CT) and support vector machine (SVM), were used to classify the Raman spectral data. The results show that linear discriminant analysis and the k-nearest neighbor model achieved a recognition accuracy of 97.62%. The method has several advantages, including no damage to bacterial samples, simple operation, and ideal identification effect. It provides guidance and a basis for accurate identification of Gram-negative and Gram-positive bacteria.

Photoacoustic spectroscopy (PAS) is a highly sensitive sensor for trace gases and is widely used. It is a research hotspot for detecting chemical poisons and toxic gases. A spectrum measurement platform based on quantum cascade laser (QCL) and a temperature-controlled photoacoustic cell were designed and constructed. Based on this, the photoacoustic spectroscopy method was established to test objects with distinct physical and chemical properties, using dimethyl methyl phosphate (DMMP) and dichloromethane (CH₂Cl₂) as research subjects. Photoacoustic spectroscopy has been successfully used to detect chemical agents in a wide concentration range of DMMP. The concentration range of 40 mg/m3 ~ 787.05 mg/m³ exhibits strong absorption and spectral characteristics. The test results confirm that photoacoustic spectroscopy detection technology is not dependent on detection wavelength, unlike traditional spectroscopy detection technology. This demonstrates the extensive application of this detection technology and provides a technical way for rapid on-site screening.

Atmospheric pollutant poses a direct threat to human health. The prevalent aerosolic particles in the atmosphere can carry and spread various pathogens, endotoxins and allergens, potentially giving rise to allergies and respiratory disorders. Most of the traditional methods for monitoring aerosolic pollution require sampling and analysis at fixed stations, with limited monitoring ranges. Remote monitoring approach based on light detection and ranging (LiDAR) technology has offered an alternative way. The transmission attenuation characteristics of laser in the atmosphere is one of the main factors affecting the detection performance of LiDAR. Especially in a high-humidity environment, the effective detection range is notably reduced by the low atmospheric visibility. Here, we introduce a theoretical model for a marine atmospheric pollutant detection LiDAR utilizing both fluorescence and Mie scattering techniques. The MODTRAN software is used to calculate the atmospheric transmittance in a high-humidity environment. The detection performance of the LiDAR system is subsequently simulated and analyzed under various marine atmospheric conditions ranging from coastal to offshore environments by incorporating historical data from the National Meteorological Center of China. The results presented here offer valuable insights into optimizing LiDAR technology for enhanced monitoring of marine atmospheric pollutants.

At present, most of the spectrometers used in food detection have some problems such as complex operation, large volume and high cost, which are not conducive to their popularization and application. Therefore, we have developed a small-scale Bluetooth fluorescence spectrometer which can be used for the detection of vegetable oil. The 405 nm semi-conductor laser can be integrated in the spectrometer to excite the fluorescence spectrum of vegetable oil, the spectrum data detected by the spectrometer is transmitted to the Android mobile phone by wireless Bluetooth, and the data is received, displayed and processed by the Android mobile phone software developed by App Inventor. The software can also be used to identify vegetable oils in real time by using machine learning to analyze spectral data on Baidu's AI open platform. In the experiment, the spectrometer collected 600 sets of spectral data of six kinds of vegetable oils, and trained and tested these data by software. The accuracy of oil identification can reach 96.1% . The experimental results show that the spectrometer can identify the kinds of vegetable oil quickly and accurately by using the developed software. It has the advantages of simple operation, high sensitivity, low cost and no pollution to the sample, it has a good application prospect in the field of food safety rapid detection.

DNA methylation is one of the earliest known modification pathways, and it regulates gene expression, which in turn influences many biological processes. Direct, label-free detection of DNA methylation with high sensitivity remains a great challenge. Surface-enhanced Raman scattering (SERS), a non-invasive and label-free vibration spectroscopy technique, offers sensitive intrinsic chemical information that makes it an attractive option for DNA analysis. In this study, we employed iodide-modified silver nanoparticles to generate highly consistent SERS signals of DNA at micromolar concentrations in aqueous solutions. This enabled the acquisition of single-base sensitive DNA fingerprint details pertaining to base methylations (such as 5-methylcytosine). As a proof of concept, the SER spectra of a DNA and its methylated counterpart were compared and analyzed, resulting in an obvious identification of DNA methylation. In particular, we first designed two DNA sequences, the sole change being that one of the nucleotides, cytosine, is replaced with a 5-methylcytosine. Further SERS experimental study revealed that methylating a single cytosine in the DNA strand caused a subtle but evident alteration in the SERS spectrum. A new Raman peak emerged at 760 cm^-1 the Raman peak at 792 cm^-1 moved to 790 cm^-1, and there was a noticeable drop in the peak intensity ratio between 1572 cm^-1 and 1634 cm^-1. This approach may provide a novel and easy-to-use tool for the label-free identification of single-molecule DNA mutations or modifications, furthering the advancement of ultra-sensitive genomic research in the future.

China is rich in coal resources, and the detection of carbon content in coal in different mining areas can also meet the requirements of coal quality in various industries. At the same time, due to the different coal-forming years and coal seam quality, the carbon content of different coal mining areas is also different, which directly affects the combustion efficiency of coal and environmental pollution emission. Laser induced breakdown spectroscopy (LIBS) has gradually become a research hotspot in soil detection due to its advantages of fast, real-time, non-destructive detection, no need for sample pretreatment, simultaneous multi-element analysis, and remote operation. However, the traditional LIBS technology has some shortcomings in detection and application, such as low sensitivity, high detection limit, poor repeatability, strong self-absorption effect, and large matrix effect, which affect its research accuracy. Therefore, this paper proposes a coal classification method based on principal component analysis (PCA) LIBS technology combined with long short-term memory network algorithm (LSTM) to classify and identify coal samples from 10 different regions. After laser ablation of coal samples and corresponding data collection, data standardization and PCA preprocessing, LSTM optimization model is used to train and continuously generate analysis network of test sets. The final results show that the accuracy of coal classification by PCA-LSTM machine learning model can reach 99.58%, which proves that LIBS technology combined with PCA-LSTM can realize fast and accurate classification of soil in different mining areas. Therefore, this method can provide a new method for coal classification in different mining areas.

This paper utilizes a Neural Radiance Fields -based method for the 3D reconstruction of hyperspectral images to enhance the 3D reconstruction effect and expand its application areas. Hyperspectral imaging technology provides rich optical information across multiple spectral bands, far exceeding traditional RGB images, and can reveal more material properties of objects. Additionally, hyperspectral images can display more texture information and detailed structures of objects, capturing more high-frequency information, making them more advantageous in 3D reconstruction. This paper proposes a NeRF-based hyperspectral image 3D reconstruction method that learns the 3D spatial density distribution and spectral information of objects through a neural network, achieving high-quality 3D image generation from any viewpoint. This study demonstrates the NeRF-based hyperspectral image 3D reconstruction method, which has broad application prospects in fields such as remote sensing, cultural heritage preservation, and agricultural monitoring. By fully utilizing hyperspectral data, the NeRF model can generate 3D images with richer details and more realistic target objects, expanding the potential for hyperspectral imaging applications in 3D reconstruction. Future research can further optimize NeRF algorithms and models, fully leveraging hyperspectral information features to improve the efficiency and accuracy of 3D reconstruction data processing, meeting more application needs and promoting the development and application of hyperspectral 3D reconstruction technology.

Calibration system is crucial for high-resolution spectrographs which are used in astronomical observation. The main scientific objective of Canary Hybrid Optical High-Resolution Ultra-Stable Spectrograph (CHORUS), under development by NIAOT, is to detect Earth-like exoplanets. The expected radial velocity (RV) precision of CHORUS is 10cm/s. Therefore, the required calibration precision is less than 10cm/s. However, the spectral intensity distributions of commonly used calibration sources vary greatly with wavelength, which reduces the signal-to-noise ratio (SNR) of calibration data. Moreover, the spectrograph detector may be saturated at some wavelengths while lacking sufficient SNR at other wavelengths. Hence, we propose a novel spectrum shaping method to improve the spectral uniformity of calibration sources and enhance the calibration precision of astronomical spectrographs. We built an experimental system based on a digital micromirror device (DMD) to demonstrate the spectral flattening of a supercontinuum source. The experimental results show that the spectral flatness is less than 0.5dB within the 585nm to738nm range.

High-precision radiometric calibration is the basis for quantitative applications of hyperspectral remote sensing. Cross-calibration facilitates the cross-comparison and radiation reference transfer between multi-source hyperspectral equipment and normalizes different remote sensors to a common radiometric baseline. In the collaborative use of different unmanned aerial vehicle (UAV) hyperspectral observations, cross-calibration helps to eliminate the differences in the radiometric and spectral scales of the multi-source remote sensors, improve the radiometric quality and interpretation consistency of the imaging from different remote sensors. However, a significant portion of the error in cross-calibration between UAV hyperspectral instruments using radiation transfer modeling comes from the assumption of aerosol type. When using the irradiance method for calculations, it is important to consider the case that the uplink radiation transfer from the UAV remote sensors passes through only a portion of the atmosphere. Therefore, cross-calibration is necessary to improve the radiation transfer model with its own characteristics. In this paper, we propose the cross-calibration method for UAV hyperspectral to address the above problems. A full set of data such as multi-gray level target images, atmospheric aerosol, water vapor content data, etc. are collected in our experiment. The method improves the traditional irradiance calibration method by combining the measured atmospheric diffuse-to-global ratio, and effectively reduces the error caused by the aerosol assumption by taking into account the special characteristics of the uplink radiation transmission path of the UAV. At the same time, considering that it is difficult to satisfy the need of cross-calibration of the whole response interval by using a single reflectance feature, the experiment adopts six kinds of targets with different gray levels for cross-calibration. Finally, the accuracy and impact of different response intervals are analyzed. The results demonstrate that the method proposed in this paper can ensure the cross-calibration accuracy more reliably, especially when the aerosol type is difficult to be determined, and it is very suitable for cross-radiometric calibration between UAV sensors.