Oil and gas leaks, along with uncontrolled energy releases, can easily trigger work safety incidents. Strengthening the visual monitoring and early warning foe oil and gas leakage is of great significance for eliminating accident risks, consolidating the foundation of safety production, and improving management efficiency. The basic theories and principles of passive Fourier transform infrared (FTIR) spectroscopy gas cloud imaging technology are firstly introduced in this paper, and then the research progress of typical products as well as the application of gas leakage imaging monitoring in the petrochemical industry both at home and abroad are reviewed and summarized. Finally, from the perspectives of standard systems, key technologies, and operational management, prospects for further promoting the application of passive FTIR spectroscopy gas cloud imaging technology in the petrochemical industry are proposed.

4J36 invar alloy is widely applied in electronic component sealing, aerospace, precision measurement, marine transportation and other fields due to its low thermal expansion coefficient. As a novel processing technique, 3D printing enters the realm of 4J26 invar alloy component fabrication. Research achievements by domestic and international scholars on 3D-printed 4J36 invar alloy are comprehensively reviewed. Findings show that SLM molding technology emerges as the primary 3D printing technology for preparing 4J36 invar alloy. The material formed by this technique exhibits a thermal expansion coefficient comparable to traditionally manufactured materials, reaching 1.8×10-6/K from 0 ℃ to 100 ℃. The density of the material can exceed 99.5%, with an ultimate tensile strength surpassing 450 MPa. Reported porosity of printing reaches 0.4%, necessitating further reduction. Overall, 3D-printed invar alloy components demonstrate potential for engineering applications. Additionally, considering the characteristics of cooled infrared detector components, it is pointed out that 3D-printed 4J36 invar alloy components hold significant promise for infrared applications.

3～5 m mid-infrared laser has attracted significant attention due to its numerous applications in gas monitoring, medical treatment and optoelectronic countermeasures. Among these, the use of 1064 nm laser-pumped MgO-doped periodically poled lithium niobate (MgO∶PPLN) as a critical method for obtaining 3～5 m mid-infrared laser, has becoming a research hotspot both domestically and internationally. In this study, to obtain high-power mid-infrared pulsed laser output, a domestically-produced widespectrum 1064 nm pulsed fiber laser is utilized to pump a multi-period MgO∶PPLN crystal in a pumping laser single-pass and signal laser single-resonance OPO (SPSR-OPO) working mode, achieving high-power mid-infrared laser outputs at wavelengths of 3181 nm, 3419 nm, 3620 nm, and 3807 nm. Under maximum pumping power of 89.7 W, the output powers are 13.29 W, 12.19 W, 9.5 W and 7.61 W, respectively, with corresponding optical-to-optical conversion efficiencies of 14.82%, 13.59%, 10.59%, and 8.48%. The study significantly lowers the technical threshold for obtaining mid-infrared lasers by employing domestically-produced broadband pulsed fiber lasers to pump MgO∶PPLN crystals, providing important experimental basis for the research and development of high-power domestically-produced mid-infrared lasers.

This paper addresses the challenge of low output power in domestically produced mid-infrared semiconductor single-bar lasers that currently limits their industrial scalability. Initially, four semiconductor target laser modules are spatially combined to increase the total output power of the quantum cascade laser in the mid-infrared band. Subsequently, through optical component design and fine-tuning of light paths, beam shaping techniques effectively resolve the degradation of beam quality during the four-channel spatial combining process, while simultaneously optimizing the beam quality post-combination. Additionally, in the drive control and temperature control module, the current drive control circuit and temperature control circuit accurately and stably control the voltage, current and temperature of the input laser, so that the quantum cascade laser can achieve stable and high-power output at room temperature. The design of quantum cascade laser with high output power at room temperature, minimal divergence angle, narrow line width and wide tunability is of great social and military importance, as well as having a wide range of practical prospects.

In large outdoor scenes, it is difficult for laser odometer to accurately register point clouds due to complex environments, resulting in a large cumulative error of motion trajectory. To solve this issue, an improved generalized iterative closest point (GICP) algorithm based on geometric and intensity constraint registration is proposed. Firstly, the average flatness and plane similarity of the point pairs to be matched are calculated by the neighborhood characteristics of the point clouds, and a geometric weight function is constructed to reduce the errors of corner points and poor corresponding point pairs during the point cloud registration process of the GICP algorithm, thereby improving the accuracy of the algorithm. Secondly, a symmetric KL (Kullback-Leibler) divergence parameter is introduced to construct intensity similarity and measure the intensity difference of point pairs to increase registration constraints. Additionally, KD-Tree is employed to accelerate the search of point cloud matching pairs to enhance the efficiency of the algorithm. The experimental results of KITTI dataset demonstrate that the average positioning error of the proposed algorithm is reduced by 58.19% and 45.64% compared with GICP and VGICP, respectively. Field experiment results show that the average positioning error of the proposed algorithm is reduced by 52.58% and 37.12% compared with GICP and VGICP, respectively, while meeting real-time requirements.

Laser paint removal technology is an environmentally friendly and safe cleaning technique. However, its application to paint removal on high-curvature cylindrical steel surfaces is associated with drawbacks, including poor paint removal uniformity and difficulties in optimizing parameters. Studying the factors influencing laser paint removal on high-curvature cylindrical steel can aid in optimizing laser parameters and improving the surface quality of cylindrical steel. In this paper, a 1064 nm laser is employed to remove paint removal from cylindrical steel aircraft engine mounts. The painted surfaces are analyzed in terms of their two-dimensional and three-dimensional morphology, roughness, and surface elements, and hardness tests are conducted on different paint removal areas to explore the influence of curvature-induced laser defocusing and changes in the incidence angle on the paint removal effectiveness of cylindrical steel. The results show that when the laser power is 500 W and other parameters are fixed, changes in defocus amount and incident angle occur as the spot moves circumferentially around the vertex of a 25 mm diameter cylindrical steel. The surface roughness decreases from 8.44 m to 7.32 m, a reduction of 13.27%. The cross-sectional hardness decreases from 254.2 HV to 221.0 HV, a reduction of 13.06%. The degree of molten state on the paint removal surface gradually decreases and approaches the original surface of the substrate. During the paint removal process, the defocused energy density caused by the displacement of the laser spot only decreases by 1.43%, indicating that the laser incidence angle is the main factor affecting the paint removal effectiveness on high-curvature cylindrical steel.

Aiming at the problems of point cloud motion distortion and low inter frame matching accuracy in the construction process of 3D point cloud maps for mobile robots, which lead to a decrease in the map construction quality and self-positioning accuracy of mobile robots, a method for 3D point cloud map construction integrating LiDAR, inertial measurement unit (IMU), and wheel encoder is proposed to address issues, such as point cloud motion distortion and low frame-to-frame matching accuracy encountered by mobile robots during the map construction process. Initially, the IMU/tachometer pre-integration values are fused to remove motion distortion from the point cloud information collected by LiDAR, and the point cloud is restored to its true position as accurately as possible. Given the small field of view angle and non-repetitive scanning method of solid-state LiDAR sensors, an outlier removal mechanism is introduced to improve the quality of line and surface feature extraction. In order to improve the matching accuracy during inter frame matching, the fusion IMU/tachometer pre-integrated values are utilized as initial matching conditions to enhance frame-to-frame matching accuracy, and then a mobile robot hardware platform is designed and constructed, and closed-loop trajectory tests are conducted. Through closed-loop trajectory testing, the results demonstrate that the proposed method in this paper performs exceptionally well in terms of the coincidence between the start-end point coordinate system and the point cloud map, and the minimum closed-loop trajectory endpoint errors under uniform motion and acceleration/deceleration motion are 0.054 m and 0.143 m, respectively.

In 3D laser projectors, specific requirements are imposed on the projection line width, necessitating a dedicated focusing module designed to adjust this parameter through modulation of the projected laser beam's echo energy. However, environmental light interference, projection distance variations, and surface reflectivity discrepancies collectively attenuate the laser echo signals to such weak levels that automated line width adjustment becomes unfeasible due to compromised signal recognition. To address this challenge, a photovoltaic cell array is employed as the echo signal receiver, while the laser beam is modulated using a 50% duty cycle square wave to generate differential bright-dark signals. The return energy of the projected beam is amplified by the primary amplifier circuit, and then summed up one-to-one according to the cycle in the field editable logic gate array to realize the digital amplification of the differential signal. This method can effectively reduce the influence of background light interference and Gaussian noise and improve the signal quality by firstly amplifying the received signal by hardware and then amplifying the differential signal by software.

With the increasing availability of laser point cloud data, research on how to extract rich point cloud feature information has become particularly important. Existing methods primarily focus on local feature learning while neglecting the relationship between point cloud positions and their features, and failing to model global information. To address this issue, an Adaptive Spatial Feature (ASF) module and the Multi-Scale Dilated Residual Block are proposed in this paper. The ASF consists of the Adaptive Feature Block and the Mixed Local Block, which can dynamically learn the relationship between point cloud positions and features, as well as eliminate uniform weighting. The Mixed Local Block combines local maximum feature data with local adaptive feature data to preserve local contextual details. The ASF is integrated into an encoder-decoder structure to form the ASF-Net network, and GroupFormer is introduced to extract global point cloud feature information. Experimental results demonstrate that ASF-Net exhibits outstanding semantic segmentation performance on the S3DIS and ScanNet v2 datasets, improving the accuracy of point cloud feature extraction.

Future advanced infrared detection sensing systems primarily aim to enhance the light field dimensional characteristics, detection target distance, spatial resolution and other performance. On-chip intelligent detection requires the adoption of specialized image processing algorithms to fuse detection data into reconstructed imagery, with readout circuit on-chip integration of intelligent autonomous identification and control algorithms circuit. Mature implementation path for intelligent control functions is established in the digital domain, where digitized readout circuit digital design constitutes the foundational prerequisite for intelligence. Data exchange between pixel arrays and algorithmic processing modules is conducted through digital signals equipped with logical processing capabilities. In this paper, a digital pixel readout circuit incorporating logic processing and storage functionalities is designed and implemented, which cooperates with image processing algorithm function to realize intelligent on-chip blind element substitution, non-uniformity correction, pixel reconstruction and motion target trajectory detection.

Infrared focal plane chips primarily consist of infrared focal plane arrays and readout integrated circuits, with flip-chip interconnection technology serving as the critical supporting technique in their manufacturing process. By introducing electrical testing to the chips after flip-chip bonding, it can provide an effective means of monitoring the interconnection quality. To meet the requirements of rapid and efficient detection of flip-chip process, a batch automatic testing method for chip interconnection connectivity is developed, realizing automatic identification of defective interconnections in test result images. Additionally, this detection method is applied after subsequent processes such as glue filling and backside thinning, with test results enabling monitoring of the influence of subsequent processes after flip-chip. The research presented in this paper satisfies the demand for improved testing efficiency while offering solutions for pinpointing process issues.

The signal of IRFPA is an important performance parameter, influenced by various factors. In this paper, based on the Planck’s law and considering multiple influencing factors, the signal value of a specific infrared detector is calculated. Through computation, a strong correlation is identified between the thickness of the chip and the signal of the IRFPA, which is subsequently verified experimentally. During verification, it is observed that when the chip thickness decreases, the signal value decreases to some extent, showing discrepancies with the calculated results. An analysis of these differences is conducted, and both analytical and experimental results reveal that the thinning thickness and thinning process of the detector chip exert significant impacts on detector performance. After secondary verification, the results substantially align with the analytical predictions.

To solve the difficult problem of large thermal response time, low absorptivity in small microbolometer pixel, a new pixel structure is proposed and designed, with small width bridge leg, quick response time and high absorptivity. By meta-surface and partial umbrella structure, the performance of high absorptivity and detective sensitivity is achieved. The thermal response time is reduced by partial umbrella design, which is theoretically calculated and simulated. Finally, a microbolometer pixel is acquired, with average absorptivity 96.8%, NETD 55 mK and thermal response time 3.8 ms, based on 8~16 m. Thesynthesis performance is approximately 209 mK·ms.

Due to its anisotropy and laminated structure, carbon fiber reinforced polymer materials make defect detection and identification difficult. Therefore, nondestructive testing research on such materials has consistently been a focal point of attention both domestically and internationally. Based on the crack heat generation mechanism, the principles of frictional heating and viscoelastic heating are primarily analyzed, revealing that heat generation at the defect is mainly related to the normal pressure and amplitude velocity. By establishing a heat generation model of a through crack in a CFRP, the finite element analysis method is employed to simulate and solve different detection parameters. The simulation results show that the greater the excitation amplitude and preload, the more obvious the heat generation phenomenon. The ultrasonic infrared thermal imaging detection platform is utilized to detect the carbon fiber reinforced composite plate with cracks. The experimental results show that under controlled variable conditions, when the excitation voltage is 180 V, the maximum temperature rise at the defect is 13.3 ℃. And when the preload is 40 N, the maximum temperature rise at the defect is 10.83 ℃. At this point, the heat generation at the crack is evident, yielding optimal detection results. The experimental results are consistent with the simulation results, which verifies that low-power ultrasonic infrared thermal imaging detection technology is a feasible and efficient method.

Atmospheric infrared radiation transmittance is the basis for infrared radiation characteristic measurement and infrared radiation inversion calculation. The atmospheric horizontal transmittance at long distance near the sea surface is greatly affected by factors such as environmental meteorological conditions and horizontal transmission distance. Based on the independently developed offshore target infrared radiation characteristic measurement system and atmospheric radiation transmission measurement system, this paper uses the sea surface sawing method to conduct long-distance medium-wave infrared measurement experiments, and uses the radiation transmission software CART to calculate the atmospheric transmittance to realize the correction of the atmospheric transmittance of long-distance medium-wave radiation transmission. This study has certain reference significance for the measurement of offshore long-distance target characteristics.

In the evolution of modern warfare, the swarm warfare system increasingly highlights its key significance, and as the indispensable‘eyes’of the system, the development of optoelectronic payloads has become a research focus in the field of military technology. In this paper, the multi-dimensional application of optoelectronic payloads in swarm warfare system is systematically elaborated, the significant differences between optoelectronic payloads and traditional optoelectronic payloads in terms of function, time statistic, imaging, signal processing and hardware platforms are deeply analyzed, and a comprehensive overview of current international advancements in swarm optoelectronic payloads is provided. Based on the above research, an innovative swarm optoelectronic payload architecture is proposed, leveraging intelligent optoelectronic networking and encompassing system components and operational modes. The technical development trend of swarm photoelectric payloads in terms of networking, integration, intelligence, imaging dimension and performance usage is further discussed, aiming to provide theoretical support and practical reference for the optimization and upgrading of photoelectric payloads in future swarm combat scenarios, driving continuous innovation and advancement in military reconnaissance technologies.

The semi-physical simulation of the guidance and control system has become an essential means in the missile development process, as it maximizes the replication of real flight environment characteristics in laboratory settings and enables comprehensive evaluation of the overall performance of missile guidance and control systems. With the accelerated development progress, the construction of the semi-physical simulation system also needs to be rapidly established. In the face of various non-standardized product requirements, it is of great significance to utilize generalized equipment resource modules for the rapid construction of HILS systems.

A novel flexible composite support structure based on adhesive-metal-adhesive layers is proposed to address the market demands for mass production of remote-sensing cameras. This structure aims to mitigate challenges such as tight assembly tolerances, high manufacturing precision requirements, and the sensitivity of mirror surface quality to assembly stresses and environmental factors (mechanical and thermal). Initially, high-stiffness structural adhesive bonding of metal components post-curing replaces traditional screw connections, with an embedded cylindrical adhesive joint is introduced to enhance assembly tolerances. Subsequently, low-elasticity silicon rubber post-curing is applied to the mirror's lateral surface, accompanied by the introduction of a metal flexible element based on series-parallel plate spring units. Additionally, a metallic flexible element, combining series and parallel plate spring units, is incorporated. The flexibility of the metallic plate springs and silicon rubber reduces assembly-induced stress and environmental disturbances on the mirror surface. Based on compliance analysis theory, the physical compliance model and its relationship with structural parameters are derived and analyzed, with a set of parameters designed for application in a mass-produced camera's mirror support. Comparative studies are conducted using compliance analysis, finite element simulations, and experimental tests on gravity-induced deformation and modal characteristic frequencies. Results indicate that the rigid-body displacement error under gravitational load, calculated using the compliance model, achieves an accuracy better than 6.1%, while the characteristic frequency error calculated using compliance matrix engineering formulas achieves an accuracy better than 10%. Finally, thermal deformation tests show that the mirror surface quality remains within 0.011 (=632.8 nm) after a 4 ℃-temperature variation. The composite flexible support structure demonstrates high adaptability to large assembly tolerances, moderate manufacturing precision, and complex mechanical and thermal environments. It has been successfully applied in the mass production of remote sensing cameras, exhibiting excellent in-orbit performance. This approach provides significant practical engineering value for future applications.

In order to obtain a lightweight primary mirror that meets the engineering requirements, the large-diameter SiC primary mirror of an optical measurement system is taken as the optimization target, and the support structure of the primary mirror is firstly determined, and then the structural dimensions of the primary mirror are optimized, and its thickness is optimized by grouping the reinforcement bars of the primary mirror, while the thickness of the front and rear panels as well as the total thickness of the primary mirror are optimized. Subsequently, the structure of the primary mirror is optimized by proposing a structural form in which triangular holes are used in conjunction with fan-shaped holes, and the face shape of the primary mirror is greatly improved. The mass of the final primary mirror is reduced by 16.7 kg compared with the initial primary mirror, the diameter-to-thickness ratio is increased from 11.4 to 12.8, and the static and dynamic characteristics meet the requirements. The influence of axial and radial temperature difference on the optical system is analyzed, and the modulation transfer function of the optical system is used as an evaluation criterion to determine the range in which the axial and radial temperature difference of the primary mirror should be controlled. Finally, the surface shape accuracy of the primary mirror is examined by a four-dimensional interferometer to verify the correctness of the finite element analysis.

JPEG Pleno compression can introduce artifacts in the compressed 4D light field images, which not only degrade their visual quality but also affect the performance of other image processing tasks. At present, there is a lack of research on JPEG Pleno light field coding distortion repair and JPEG Pleno coding distortion light field dataset for network training. In this paper, an end-to-end flexible blind convolutional neural network, namely JPLARNet, is proposed, which fully considers the characteristics of JPEG Pleno encoding and can predict the coding quality factor to control the trade-off between artifact removal and detail preservation. Specifically, JPLARNet initially employs a spatial-angle decoupling module to perform preliminary feature extraction on light field images. The extracted features are then processed by a multi-scale decoupler to obtain predicted compression factors and high-level semantic features. Subsequently, a compression factor fusion module is utilized to embed the predicted compression factors into the subsequent reconstructor module, and then embeds the predicted compression factors into the subsequent reconstruction module through the compression factor fusion module, thereby guiding artifact removal in compressed light field images. In addition, two modules, namely the LK Down-sample module and the mixed attention enhancement module, are constructed for downsampling and image enhancement of the reconstructed images, respectively. The experimental results show that on the six compression qualities of the constructed JPL-DATA dataset with compression factors ranging from 5000 to 200000, the average gain in YUV-PSNR/Y-SSIM of the light field image after artifact removal compared to before removal is 0.81 dB/0.025. By taking into account the characteristics of JPEG Pleno encoded light field image, the proposed method achieves light field images with higher subjective and objective quality than the JPEG artifact removal method of 2D images.

In the registration of point clouds with low overlap rate, traditional methods often struggle with features and matching difficulties, frequently falling into local optimum under large pose errors or complex transformation scenarios, which compromises registration accuracy. To solve these problems, an adaptive graph convolution model with progressive feature fusion pyramid network is designed to establish correspondences between point clouds from coarse to fine scales. Firstly, Adaptive Graph Convolution (AGConv) is employed to extract and encode spatial features, and then Progressive Feature Pyramid Network (AFPN) is utilized to fuse semantic information across multiple scales to jointly enhance the performance of the model in complex 3D scene understanding and analysis tasks. Secondly, a geometric Transformer is introduced to strengthen the model's comprehension of global structures and correlations, and achieve high-quality super-point matching. Finally, a local-to-global registration method is designed by integrating AGConv and AFPN, leveraging learned local point features from the backbone network and resolving global ambiguity problems through superposition point matching, thereby improving the robustness of the algorithm. Experiments show that the proposed network significantly improves the registration accuracy of point clouds with low overlap rate.

CdZnTe crystal material is the preferred substrate material for third-generation high-performance, long/very long wave cadmium telluride infrared focal plane detectors. However, due to the inherent characteristics of CdZnTe material, the presence of polycrystalline and twin regions in the grown crystals affects device performance. At present, the main method for cutting single crystal wafers is through manual identification of single crystal regions, resulting in low efficiency and unclear contour recognition. In this paper, based on a multi-angle wafer surface topography visual recognition device, neural network-based image segmentation technology has been applied to the identification of single-crystal regions in Te-Zn-Cd wafers, enabling the automatic differentiation between Te-Zn-Cd single-crystal and polycrystalline regions. This provides a foundation for the automatic cutting process of unit price regions.

Aiming at the issues of high onboard core computing requirements and low computational efficiency in UAV multi-target recognition neural networks, this paper proposes a lightweight neural network designed to reduce the algorithm's reliance on hardware memory and to meet the demands of unmanned equipment for synchronized, lightweight, and highly efficient neural networks. A multi-channel splitting convolutional computing strategy is introduced to decrease the amount of serial computation. A dual allocation strategy is employed to enhance the capability of dense target selection and to reduce the inference process. Angular loss is utilized to address the misalignment between prediction frames, thereby improving the model's inference accuracy and convergence rate. Experimental results demonstrate that the proposed algorithm achieves a recognition accuracy of 94.3%. Compared with the mainstream models YOLOV5n and YOLOV8s on HIT-UAV data set, the experiment shows that the recognition rate of small targets is 2.8% and 0.9% higher. The recognition rate of medium target is 2.2% and 1.3% higher. Enabling UAVs to possess efficient and precise end-to-ground target recognition capabilities even with limited computational resources.

In this paper, a sensor capable of simultaneously measuring refractive index of two kinds of media is proposed. The sensor consists of a single-layer graphene, a thin dielectric layer, a dielectric cavity A and a multilayer photonic crystal with a defect layer (i.e. a dielectric cavity B). When the incident light is introduced from the graphene side, the terahertz Tamm plasmon-polaritons (TPPs) mode at the interface between graphene and thin dielectric layer and the defect mode in photonic crystal are simultaneously excited. These two modes interact and couple to form an asymmetric Fano resonance absorption spectrum. The absorption peak wavelength coincides with the excitation wavelength of TPPs mode and is exclusively sensitive to the refractive index of the medium in cavity A, while the absorption Valley wavelength matches the defective mode wavelength and is solely sensitive to the refractive index of the medium in the defective layer. Therefore, the simultaneous detection of the refractive index of the two media can be realized. Numerical simulation results demonstrate that the sensor achieves refractive index measurement ranges of 1.3~1.46 and 1~1.005 for media in cavities A and B, respectively, with corresponding sensitivities of 0.125 THz/RIU and 0.6 THz/RIU. The study in this paper provides novel approach for simultaneous multi-medium refractive index measurement by leveraging coupling between TPPs modes and distinct modes.