The fingerprint absorption spectrum of mid-infrared fundamental frequency is characterized by strong absorption, broad and dense spectral lines. In general, as molecules absorbing infrared spectra have mid-infrared base-frequency absorption intensity near 3000 cm-1 that is about two orders of magnitude higher than that of the near-infrared absorption, and thus it has gradually become a research hotspot for cavity ring-down spectroscopy (CRDS) technology. In this paper, the working principle and technical advantages of CRDS technology are briefly described, and the characteristics of mid-infrared CRDS technology are introduced. Moreover, the differences between near infrared and mid-infrared CRDS technology are analyzed and compared. The research status of CRDS technology based on mid-infrared band is discussed finally.

Infrared precision-guided weapons, typified by man-portable air defence systems (MANPADS), pose a significant threat to military and civilian aircraft. At present, the most effective countermeasure against heat-seeking guided weapon is the directional infrared jamming system, which uses infrared-band lasers as the light source. This paper mainly introduces how Raytheon Company develops a new generation of directed infrared countermeasure system based on the existing AIM-9X seeker technology. The system has many advantages, such as high efficiency, reliability, compactness, light weight and low price, and through the expansion of the functions, it can easily cope with the threat of missiles in various wavelength bands from ultraviolet to mid-infrared.

As is well known, thermal effect is a major bottleneck limiting the development of high-power and high-energy lasers. During the process of producing high-energy lasers, a large amount of waste heat is generated, which affects the beam quality of high-energy lasers and even their normal operation. In order to ensure the stable operation of the high-energy laser and study the heat distribution state of its working material in the process of heat dissipation, this paper establishes a dual end water entry microchannel heat dissipation model for high-energy Zig-Zag Flat noodles laser amplifier. The heat dissipation of microchannel and cavity heat sink is compared under rated conditions using CFD simulation software. The variable parameters of the model are also studied: channel height, fin thickness, And the impact of water flow rate on heat dissipation performance. Simulation studies have found that the cooling effect of the microchannel heat sink proposed in this article is better than that of full cavity water cooling. The microchannel heat sink controls the maximum temperature difference on the crystal surface within 4 ℃, and the surface temperature is also reduced by 32%; Optimizing channel parameters within the allowable range of pressure drop can further improve the cooling effect by 10%, achieving distributed and efficient heat dissipation of gain media.

In this paper, the discrete dipole approximation method is adopted to study the scattering characteristics of dust particles in different Chebyshev models in order to investigate the scattering characteristics of dust particles in the terahertz band. The effects of different physical properties such as particle size and particle shape on their scattering characteristics including backscatter section (Cbak), attenuation section (Cext), scattering section (Csca), single-scatter albedo () for the frequencies of 1.5 THz, 2.524 THz, and 3.437 THz are calculated, respectively. The results show that under the same ripple parameters, the particle size at 1.5 THz has a greater effect on Cext and Csca; the particle size changes at 2.524 THz has a larger effect on Cbak. Under the same frequency, the particle size change of the waveform parameter of 2~5 has a significantly different effect on each optical cross-section compared with the other shapes. Under the three frequencies, particles of each ripple parameter with the effective radius larger than 10 m when the single scattering albedo is greater than 0.95.

To improve the accuracy of aircraft wake detection and inversion based on LiDAR, an estimation method for atmospheric background turbulence dissipation rate (EDR) on the basis of Kolmogorov structure function is established according to the scanned radial wind speed data. Then, on the strength of a probabilistic two-phase wake vortex Decay and Transport Model, the influence of turbulence on the wake dissipation process is taken into account to achieve the prediction of wake intensity circulation and vortex core motion trend based on historical detection data. By combining environmental detection data with predictive models, the inversion accuracy of wake characteristic parameters is improved. The research is shown that tail vortex trajectories predicted using the model in this paper are 59.5% and 64.8% more accurate in terms of radial distance and angle, respectively compared to the inverse algorithm.

In this paper, a multi-modal combination imaging method based on synthetic aperture focusing technique is proposed to address the laser ultrasonic imaging problem of internal fatigue cracks in titanium alloy. The scanning laser source method is employed to extract time-domain mode-echo signals from different detection positions. By analyzing the amplitude distribution of reflected longitudinal waves and reflected shear waves in the B-scan image, the temporal characteristics of crack echo signals are partitioned and recombined. Furthermore, inversion focusing is applied to accomplish the recognition and image reconstruction of internal cracks. The feasibility of this approach is demonstrated by conducting numerical simulations of the internal crack imaging process in titanium alloy plates using COMSOL Multiphysics, and its viability is further validated using MATLAB image algorithms. Compared to single-mode wave imaging, the imaging method that combines reflected longitudinal wave and reflected shear waves effectively reduces artifacts, enabling accurate localization and image reconstruction of internal cracks. The relative errors of the reconstructed crack positions in the X-axis and Y-axis are both within 3%.

As a cutting-edge laser detection technology, single photon laser ranging technology has been successfully applied in the fields of lunar ranging, satellite ranging, and ground measurement. However, single photon ranging still needs to be solved when tracking and ranging high-speed moving targets on airborne air-to-air and ground-to-air platforms, where the echo photons fall in different time windows, resulting in direct counting that cannot effectively extract the signals. Aiming at the application requirements of single photon laser ranging under air-to-air conditions, a single photon ranging method is designed based on time-dependent photon counting technology for high-speed moving targets under full-time, wide-time domain, and multi-noise conditions. The laser echo photon signals are extracted using arrayed single photon detectors and the adjacent time window correlation statistical multi frame processing algorithm, and simulation experiments are carried out on the Matlab platform. The echo photon signals are extracted under the conditions that the maximum range is more than 100 km, the background noise count rate is about 5 MHz, and the average value of the single pulse echo photon count is 1 using the multivariate array single-photon detector. This method can overcome the limitations of traditional single-photon detection, which can only be aligned to static target ranging, and can only be applied in weak background noise and predictable target trajectory conditions such as small receiving field of view and wavelet gate range. The single photon detection can be extended from the range finding of static targets at night on a fixed platform to the range finding of high-speed moving targets all day on a general platform.

In this paper, a feature matching method for LiDAR based on Kalman fusion is proposed to address the problem of inaccurate localization using line segment features in existing LiDAR feature matching algorithms. Firstly, a frame of LiDAR data is scanned, and local map is generated by using an improved method for extracting line segment features. The rotation and translation parameters of the partial map are then determined, and the partial map is matched with the global map to obtain the matching result according to the relative deviation. Then, based on the Kalman filter, the IMU data is used to predict the estimation for the next moment, and the LiDAR matching result is used as the observation. Finally, the two results are fused to obtain the optimal estimation. The experimental results show that this method is more accurate in matching line segment features compared to the existing feature matching algorithms, which leads to better precision and robustness in localization and navigation.

In coherent LiDAR remote ranging, in order to improve the detection range and range resolution of the LiDAR, a large time broadband product signal such as a frequency modulation signal is usually used to modulate the optical carrier, and pulse compression processing is performed at the receiving end. To reduce the data required for digital signal processing at the receiving end and improve the real-time performance of the calculation, it is necessary to down-convert the received signal to an appropriate frequency band. The traditional heterodyne radar receiver requires a single down conversion of the optical and electrical signals, resulting in a more complex system structure and by the limitations of the device non-idealisation, additional noise is introduced during the downconversion process. In addition, there is the problem of image frequency noise interference, which leads to the degradation of the SNR of the demodulated signal. In this paper, a scheme is proposed to downconvert the optical signals to the frequency band required for pulse compression, which is carried out using orthogonal demodulation to simplify the system structure and suppress mirror frequency noise. Firstly, the local oscillator light is frequency shifted and divided into two beams, and the two local oscillators are made orthogonal to each other by controlling the phase. The signal light is divided into two beams and mixed with two local oscillators on the surface of the photodetector. Then, the electrical signals are collected and the amplitude-phase imbalance is corrected through relevant algorithms. Through simulation and experiments, this method can effectively simplify the structure of the laser coherent radar receiver system while avoiding mirror frequency noise interference. At a sampling rate of 10 GSps, the SNR of demodulated signal is improved by about 3 dB compared to a heterodyne receiver.

Aiming at the reflection intensity characteristics of road markings from vehicle-mounted laser point cloud, an urban road marking extraction method based on vehicle laser point cloud is proposed. Firstly, a ground filtering method combining cloth simulation filtering and elevation skewness balance is presented to take advantage of the adaptability of skewness balance filtering to eliminate the problem of low vegetation remaining after cloth filtering. Then, the normal vector density clustering is adopted to extract the road point cloud, and the road point cloud is converted into an intensity feature map by inverse distance weighted interpolation. In order to alleviate the jaggedness extracted from the markers, fast bootstrap filtering is introduced to smooth the edge information of the road markers. Finally, the maximum entropy threshold segmentation and morphological ratio filtering are used to refine the road marker. The experimental results show that the method can effectively extract the road marking point cloud, the average recall rate of extraction is 80.98%, the average accuracy rate is 96.89%, and the average comprehensive evaluation index is 88.19% and it can use the road marking point cloud intensity information to extract the road marking point cloud in a more complete way.

In this paper, the structural design and assembly process of continuous zoom thermal imager are studied. Vibration and shock simulation analysis are carried out at the design stage. The optical-mechanical assembly process of continuous zoom module is proposed. The measurement method of off-axis degree between narrow field and assembly reference is built. The optical axis consistency test and optical axis stabilization test are finished. The results of testing and experimental verification show that the key mechanical performance indicators are excellent, and prove that the system structure of the thermal imager is designed to meet the needs of engineering applications.

With the development of linear infrared detector technology, the demand for ultra large field of view scanning is an important direction for application of linear infrared detectors. Ultra large field of view scanning is usually accomplished by splicing of multiple linear infrared detectors, and the core of the splicing design is mainly the packaging structure design. How to fully lead the detector signal completely out of the packaging body is what needs to be done in the electrical design of packaging structure design. The article introduces the packaging electrical design of a splicing structure of a linear infrared detector. Firstly, the splicing method of the splicing structure is described, including the splicing method of the spliced structure, as well as the electrical structural design scheme to meet the structure, especially for the secondary processing part of the signal. Finally, the electrical wiring design solution for the electrical structure scheme is presented.

In practical engineering applications, the short duration of rolling bearing fault states leads to imbalanced datasets, making it difficult to use deep learning algorithms for fault diagnosis. In this paper, a n infrared diagnosis method for rolling bearing faults based on the combination of the Wasserstein distance-based gradient penalty generative adversarial network (WGAN-GP) and a support vector machine-based convolutional neural network (CNN-SVM) is proposed. The imbalanced dataset is constructed from infrared thermal images, and WGAN-GP is used to augment the imbalanced data to achieve dataset balance, after which the CNN-SVM model is then applied to the dataset to extract deep features and complete fault classification. The experimental results show that the model combining WGAN-GP with CNN-SVM performs well under imbalanced datasets, with better fault diagnosis capability compared to other models, and reduces the time spent in the fault classification stage by more than 16.89%.

In response to the demand for accurate assessment of infrared thermal fault characteristics of power equipment, a multi-feature aggregated characterization of circuit breaker thermal fault diagnosis rating method is proposed, and the data test is carried out using infrared images of high-voltage circuit breakers as examples. Firstly, on the basis of the background separation of high-voltage circuit breaker infrared images, the equipment is accurately divided into regions to extract the temperature information of each region. Secondly, the Mean-shift and the improved region growth method are applied to fuse and accurately extract the area of the fault heat-emitting region. Then, a multi-dimensional aggregated characterization matrix is designed to combine the heat-emitting area, hot spot temperature, hot spot temperature difference, heat-emitting location, temperature rise of two identical positions of the same equipment and other eigenvalues into a multi-feature vector matrix, and the on-site case data is adopted to construct a correlation library of this vector matrix and HV circuit breaker fault types, levels and treatment opinions. Finally, 1002 sets of multi-feature vectors from 350 infrared images of high-voltage circuit breakers are trained and tested. The results show that the F-measure and Kappa coefficients of the multi-feature vector data extracted by this method using GWO-SVM classifier test are 96% and 95.43%, respectively, which can achieve the all-types of diagnostic rating and accurate localization of thermal faults in high-voltage circuit breaker equipment.

To address the problem that infrared video lacks texture detail features which is difficult to balance the computational complexity and recognition accuracy in human behavior recognition, a global bilinear attention-based behavior recognition method for infrared video is proposed in this paper. Firstly, in order to efficiently compute human behavior in infrared video, a joint extraction module based on a two-stage detection network is designed to obtain human joint point information, and the resulting 3D heat map of joints is innovatively used as an input feature for the human behaviour recognition network in infrared video. Moreover, to further improve the recognition accuracy on the basis of lightweight computation, a global bilinear attention-based 3D convolutional network is proposed to enhance the attention from both spatial and channel dimensions modeling capability to capture global structural information. The experimental results on the InfAR and IITR-IAR datasets demonstrate the effectiveness of the method in infrared video behavior recognition.

In this paper, a multi-channel spectral detection system is designed to address the need for multi-spectral radiometric temperature measurement of high temperature targets over long distances. The front end of the system uses a bi-reflective Cassegrain structure to collect the spectral radiation of the target, while the back end includes a long-wave infrared imaging targeting system for target location and a dual-band spectral system operating at 380~1050 nm and 980~2550 nm respectively. The spectral system adopts Czerny-Turner structure, with a resolution better than 5 nm in the 380~1050 nm band and 15 nm in the 980~2550 nm band, and the number of spectral channels is greater than 220. The 400 m target outfield experiments and the 700~2300 ℃ high temperature blackbody temperature experiments show that the multi-channel spectral detection system is capable of targeting and spectral detection of long-range targets. The multi-channel spectral detection system designed in this paper is expected to be used for long-range temperature detection on the ground, in aviation and in space.

In this paper, analytical formulas of M2 factor and angular extension (z) of partially coherent twisted vortex beam (PCTVB) propagating in oceanic turbulence are derived based on the extended Huygens-Fresnel principle and the Wigner distribution function (WDF) second moment. The effects of oceanic turbulence on M2 factor and angular spread (z) of PCTVB are studied in detail by numerical simulation methods, and the results show that the M2 factor and (z) of PCTVB is greatly affected by oceanic turbulence with the large rate of dissipation of mean-square temperature and large relative strength of temperature and salinity undulations as well as the small rate of dissipation of turbulent kinetic energy per unit mass of fluid and small anisotropic factor. In addition, it is found that the PCTVB has better resistance to ocean turbulence compared to the non-twisted vortex beam (PCVB), and the M2 factor and angular extension (z) of the PCTVB are significantly reduced and the beam's resistance to ocean turbulence is enhanced by increasing the topological charge m as well as the absolute value of the distortion factor ||. And increasing the beam waist width w0 and wavelength as well as decreasing the initial coherence lengths (=x,y) can likewise increase the beam's resistance to ocean turbulence. The numerical results in this paper are of great significance for ocean optical communications.

Currently, the applications of convolutional neural networks to the task of fusing infrared and visible images have achieved better fusion results. Many of these methods are based on network models with self-encoder architectures, which are trained in a self-supervised methods and require the use of hand-designed fusion strategies to fuse features in the testing phase. However, existing methods based on self-encoder networks rarely make full use of both shallow and deep features, and convolutional neural networks are limited by the receptive field, making it more difficult to establish long-range dependencies and thus losing global information. In contrast, Transformer, with the help of self-attention mechanism, can establish long-range dependencies and effectively obtain global contextual information. In terms of fusion strategies, most of the methods are designed in a crude way and do not specifically consider the characteristics of different modal images. Therefore, CNN and Transformer are combined in the encoder to enable the encoder to extract more comprehensive features. And the attention model is applied to the fusion strategy to optimize the features in a more refined way. The experimental results show that the fusion algorithm achieves excellent results in both subjective and objective evaluations compared to other image fusion algorithms.

With the application of deep learning in computer vision, its large amount of data, complex network layer structure, insufficient resources in hardware deployment and high delay have become key problems. This paper, by analyzing the advantages and disadvantages of five representative lightweight networks, proposes a lightweight network improvement based on MobileNet, which applies lightweight networks to infrared target detection field. FPGA is used as the hardware carrier. In this network, Tanh activation function is used to replace the original activation function and the number of network layers is simplified to adapt to the feature extraction of infrared targets. In view of the problems existing in the hardware implementation of deep learning target detection algorithm, such as large amount of data, large resource occupation and high calculation delay, FPGA is adopted for hardware implementation. The experiment shows that on Xilinx Zynq-7020 XA development board, the clock frequency is set to 100 MHz and the input image size is 640×512. The improved MobileNet can achieve each image of 5.1 ms with the same accuracy as the original one.

In response to the current demand for monitoring the waterlogging events and the extent of waterlogging in tunnels, two new types of tunnel waterlogging sensors are developed, and a distributed sensing method for tunnel waterlogging based on distributed temperature sensing (DTS) is proposed in this paper. Firstly, the spatial resolution of the sensor is improved by encapsulating the optical sensing cable in a two-dimensional plane spiral pattern and an actively heated method is used to significantly improve the temperature difference of the optical sensing cable at the water-air interface, thereby realizing the identification of the events and range of tunnel waterlogging. Then, through the indoor model tests, the performance and influencing factors of the optical sensing cable are studied. The results show that the two sensors developed have higher spatial resolution. Due to the smaller spacing of the spiral circle, the spatial resolution and positioning accuracy of the spiral type I sensor are higher than those of the spiral type II sensor. Both sensors can accurately identify the water-air interface under various typical working conditions, and their positioning errors are less than 10 cm. The positioning accuracy of the sensors can be improved by appropriately increasing the heating temperature, and the positioning accuracy under the heating conditions of 50 ℃ is significantly better than 35 ℃ and 20 ℃.