Abstract:The sensitivity of the detector is the core technical indicator of the infrared detector. Short-wave infrared detector has low dark current and the sensitivity will be limited by the inherent read-out circuit noise of the detection system. Therefore， it is an effective way to further enhance the sensitivity by introducing internal gain into the detector. The heterogeneous phototransistor has advantages of high gain， low operating bias， and low excess noise， which provides novel approach for high-sensitive detection. This paper mainly focuses on the simulation design of InGaAs/GaAsSb type-II superlattice short-wave infrared phototransistor， and studies the dependence of the device size on the optoelectronic characteristics. The results show that a higher gain， a lower dark current， and a faster response can be achieved by a smaller base size. Based on the optimization design of size structure， a noise equivalent photon lower than 10 can be achieved， which provides a new technical approach to achieve high-sensitive heterogeneous phototransistor detector.

Abstract:The preparation of quantum dot thin films and the accurate measurement of optical constants are particularly important for the development and application of quantum dot devices in optoelectronic fields. At present， the optical constants of MoTe₂ single-crystal films prepared by mechanical exfoliation and chemical vapor deposition are relatively mature. However， the optical constants of 2H-MoTe₂ quantum dot films are rarely reported.In this work， 2H-MoTe₂ quantum dots were prepared by ultrasonic assisted liquid phase exfoliation， and of 2H-MoTe₂ quantum dots with two sizes were acquired via adjusting the type of solvent and the order of ultrasonic process. The optical constants such as refractive index， extinction coefficient and dielectric constant of quantum dot films of two sizes were studied by B-spline model and Tauc-Lorentz model using ellipsometry. The results demonstrate that the 2H-MoTe₂ quantum dots with two sizes have similar refractive index， extinction coefficient and a wider spectral absorption in the visible to near infrared band. And compared with MoTe₂ bulk material， 2H-MoTe， quantom dots have a lower dielectric constant.

Abstract:Lasers with wavelengths around 2 μm are located in the atmospheric transmission window and at the strong absorption peak of water， and have important applications in medicine， LIDAR， material processing， and pump sources for mid-infrared lasers. The thulium-doped fiber laser （TDFL） stands out as a critical light source capable of delivering high power outputs at this wavelength. In this paper， to address the problems of relaxation oscillation and inter-modal four-wave mixing in a quasi-continuous wave （QCW） TDFL， the time and frequency-domain output characteristics of the laser are optimized by increasing the bias current， optimizing the length of the gain fiber， and changing the diameter of the fiber coiling， etc. The effects of different gain fiber structures on the fiber transmission modes are also investigated. The developed QCW-TDFL achieves a peak power of 894 W and an average power of 89.4 W at a central wavelength of 1 939.2 nm with a pulse width of 100 μs， a repetition frequency of 1 kHz and a duty cycle of 10%， and obtains stable and controllable pulse output waveform and spectral characteristics.

Abstract:As an important emitter for terahertz radiation， photoconductive antenna arrays are limited in their application flexibility due to the fixed polarization state of traditional antennas. In response to this issue， we have designed and studied a polarization-adjustable four-element terahertz gallium arsenide photoconductive antenna array， aiming to enhance its versatility and applicability in various applications. By precisely controlling the excitation of each element， the antenna array can achieve precise control of linearly and circularly polarized terahertz waves through in-phase unequal amplitude excitation and phase difference excitation. The results show that with in-phase unequal amplitude excitation， flexible control of linearly polarized terahertz waves within a 360-degree range can be achieved. With a 90-degree phase difference excitation， circularly polarized terahertz waves are generated， with a -10 dB impedance bandwidth range of 0.057 THz to 1.013 THz and a relative bandwidth of 178.69%. The axial ratio bandwidth range is 0.815 THz to 0.947 THz， with a relative bandwidth of 14.98%.

Abstract:In this paper， a linear optimization method（LOM） for the design of terahertz circuits is presented， aimed at enhancing the simulation efficacy and reducing the time of the circuit design workflow. This method enables the rapid determination of optimal embedding impedance for diodes across a specific bandwidth to achieve maximum efficiency through harmonic balance simulations. By optimizing the linear matching circuit with the optimal embedding impedance， the method effectively segregates the simulation of the linear segments from the nonlinear segments in the frequency multiplier circuit， substantially improving the speed of simulations. The design of on-chip linear matching circuits adopts a modular circuit design strategy， incorporating fixed load resistors to simplify the matching challenge. Utilizing this approach， a 340 GHz frequency doubler was developed and measured. The results demonstrate that， across a bandwidth of 330 GHz to 342 GHz， the efficiency of the doubler remains above 10%， with an input power ranging from 98 mW to 141mW and an output power exceeding 13 mW. Notably， at an input power of 141 mW， a peak output power of 21.8 mW was achieved at 334 GHz， corresponding to an efficiency of 15.8%.

Abstract:The presence of water in lunar materials can significantly impact the evolution of lunar geology and environment， as well as provide necessary conditions for the utilization of lunar resources. However， due to the limitations of lunar remote sensing methods， it is challenging to obtain direct evidence of water or determine its form of occurrence. Laser Raman spectroscopy， on the other hand， can provide valuable information on the type， distribution， and content of water in lunar materials without the need for illumination， sample pretreatment， or destructive measures. In this study， we utilized Raman spectroscopy to detect and quantify the water-containing characteristics of typical lunar rocks and minerals， including adsorbed water， ice， crystalline water， and hydroxyl-structured water. First， we used a 532 nm laser micro-Raman spectroscopy to identify and analyze the water-containing signals of various forms of water in lunar soil simulants. We then examined and analyzed the detection limits of adsorbed water， crystalline water， and hydroxyl-structured water in these simulants， as well as the relationship between their content and signal intensity. Finally， we employed linear regression （LR）， ridge regression （RR）， and partial least squares regression （PLSR） to quantitatively analyze the contents of these three forms of water in the lunar soil simulants. Our results demonstrate that the characteristic spectral peaks of the four forms of water in the lunar soil simulants can be clearly identified， with peak distribution regions located at 100-1 700 cm^-1 and 2 600-3 900 cm^-1 for the lunar soil components and water bodies， respectively. The spectral peaks of water are a combination of broad envelope peaks of hydrogen-bonded OH and sharp peaks of non-hydrogen-bonded OH stretching vibrations in varying proportions. The detection limits for adsorbed water， crystalline water （MgSO₄·7H₂O）， and hydroxyl water （Al₂Si₂O₅（OH）₄） in the lunar soil simulants are 1.3 wt%， 0.8 wt%， and 0.3 wt%， respectively. There is a linear relationship between the intensity of water-containing peaks and the water content in the lunar soil simulants， with root mean square errors of 1.75 wt%， 1.16 wt%， and 1.19 wt% obtained through LR， RR， and PLSR.

Abstract:When extracting coastal zone tidal flats using remote sensing transient images， the influence of tides greatly limits the accuracy of tidal flat spatial distribution extraction. With the purpose of weakening the influence of tides， a method of extracting coastal zone tidal flats by combining time-series Sentinel-2 images and tidal flat index was proposed. First， based on the Sentinel-2 time-series image data， we us the quantize synthesis method to generate high- and low-tide images， and then analyz the spectral reluctance characteristics of different land classes on the high- and low-tide images. A NIR-band tidal flat extraction index that excludes the interference of the tidal transient was constructed. Secondly， the image spectral information and the tidal flat extraction index were input into a machine learning algorithm to realize fast and efficient extraction of the tidal flat. In addition， the study discussed the separability of the tidal flats index and the generalizability of the methodology. The results show that the tidal flat''s extraction index constructed in this research had a good separability for tidal flats， the overall accuracy of tidal flats extraction was 93.02%， the Kappa coefficient was 0.86， and the proposed method had good applicability to remote sensing images containing near-infrared bands. This method can realize automatic and rapid tidal flat extraction， and provide data support for the sustainable management and protection of coastal zone resources.

Abstract:Urban tree species provide various essential ecosystem services in cities， such as regulating urban temperatures， reducing noise， capturing carbon， and mitigating the urban heat island effect. The quality of these services is influenced by species diversity， tree health， and the distribution and composition of trees. Traditionally， data on urban trees has been collected through field surveys and manual interpretation of remote sensing images. In this study， we evaluated the effectiveness of multispectral airborne laser scanning （ALS） data in classifying 24 common urban roadside tree species in Espoo， Finland. Tree crown structure information， intensity features， and spectral data were used for classification. Eight different machine learning algorithms were tested， with the extra trees （ET） algorithm performing the best， achieving an overall accuracy of 71.7% using multispectral LiDAR data. This result highlights that integrating structural and spectral information within a single framework can improve classification accuracy. Future research will focus on identifying the most important features for species classification and developing algorithms with greater efficiency and accuracy.

Abstract:To address the issues of low detection rate and high false alarm rate caused by complex background during sub-pixel aerial aircraft detection in hyperspectral remote sensing image， an aerial aircraft detection method was proposed based on contrails cloud proposal. Firstly， a hyperspectral semantic segmentation model was used to search for the contrails cloud， and regions of interest（ROIs） of aircraft were proposed to reduce invalid search ranges and suppress false alarms based on the contrails cloud. Secondly， an endmember extraction algorithm based on dictionary learning and semi-blind non-negative matrix factorization was proposed to improve the accuracy of aircraft endmember extraction for hyperspectral subpixels. Finally， verification experiments were carried out on the hyperspectral remote sensing image dataset of Gaofen-5 satellite. The results demonstrated that the algorithm proposed in this paper can effectively suppress false alarms in complex scenes， and significantly improve the detection rate and detection accuracy of sub-pixel aerial vehicles.

Abstract:The energy received in the mid-infrared （MIR） band at the sensor''s aperture includes both reflected solar energy and the emitted energy from Earth''s surface. Typically， the reflected solar energy in this band is weak. However， under certain conditions， such as in sun glint regions on the sea surface， the reflected solar energy detected by the MIR channel can be substantial. Currently， the application of sun glints physical models in the MIR band is not yet-clear. This study investigates the accuracy of applying different visible light and shortwave infrared sun glint models in the MIR band to evaluate their applicability. The paper selects three models，namely Breon-Henriot， Ebuchi-Kizu， to first Wu， and evaluate the sensitivity of each sun glint model. Subsequently， four selected MODIS sun glint images as data sources， and to evaluate their applicability ERA5 reanalysis data matched with satellite data wal usel to calculate atmospheric parameters. The solar radiation intensity reflected by the sea surface is computed using the three models. The accuracy of each model is then further validated with an MIR radiation transfer model. The results show that the Breon-Henriot model generally performs best in terms of correlation coefficient and root-mean-square error compared to MODIS measurements. These findings not only extend the application range of sun glint models in the MIR band but also enhance the MIR forward modeling system， providing new theoretical support for MIR radiation transfer and improving the effectiveness and accuracy of MIR remote sensing products in climate change monitoring and sea surface temperature dynamic analysis.

Abstract:The acquisition of aircraft altitude information is crucial for aviation safety and traffic control applications.Currently， it is difficult to obtain the altitude information of aircraft only through passive remote sensing methods. By combining atmospheric parameters retrieval and high-sensitivity infrared detection technology， space-borne remote sensing platforms can achieve accurate measurement of target thermal radiation information and have the potential for quantitative observation of aircraft characteristic information. A method for estimating the altitude of airborne targets based on infrared multi-channel feature matching is proposed in this paper. Firstly， a thermal infrared radiation characteristic observation model of aircraft is established， which is based on the thermal infrared radiation characteristics of large aircraft and atmospheric radiative transfer models. Secondly， based on the observation model， a spectral database of aircraft at different altitudes and flight states under different atmospheric conditions can be obtained by simulating. Thirdly， target spectral information can be extracted from remote sensing images and the altitude information can be estimated with spectral angle matching （SAM）. Finally， verification and analysis were completed using simulation data and SDGSAT-1 in-orbit data. The results indicate that the proposed method can achieve kilometer-level estimation accuracy for aircraft at cruising altitude. This method provides a new solution for estimating the altitude of aircraft and has important application potential.Key words： thermal infrared remote sensing， altitude estimate， spectral angle matching， large aircraft

Abstract:Smooth objects such as metals， optical mirrors， and silicon wafers generally have extremely low emissivity and high reflectivity， and are called low emissivity objects. The extremely weak radiation from low emissivity objects will be submerged by the environmental radiation reflected from their surfaces. Infrared temperature measurement of low emissivity objects has always been a challenge in the field of infrared temperature measurement. Due to the continuously growing demand for non-contact temperature measurement of low emissivity objects in fields such as metal smelting， solar telescope thermal control， and semiconductor production， a large number of infrared temperature measurement methods for low emissivity objects have been proposed. First， this paper elaborates on the difficulties of the infrared temperature measurement of low emissivity objects and summarizes the temperature measurement methods currently used for low emissivity objects into five categories. Then， the basic principles and technical routes of each temperature measurement method were summarized， and the advantages and disadvantages of each temperature measurement method were analyzed in detail. Finally， the possible development directions of temperature measurement for low emissivity objects were discussed.

Abstract:Space-borne infrared remote sensing images have significant applications in environmental monitoring and military reconnaissance. Nonetheless， due to technological limitations， atmospheric disturbances， and sensor noise， these images suffer from insufficient resolution and blurred texture details， severely restricting the accuracy of subsequent analysis and processing. To address these issues， a new super-resolution generative adversarial network model is proposed. This model integrates dense connections with the Swin Transformer architecture to achieve effective cross-layer feature transmission and contextual information utilization while enhancing the model''s global feature extraction capabilities. Furthermore， the traditional residual connection is improved with multi-scale channel attention-based feature fusion， allowing the network to more flexibly integrate multi-scale features， thereby enhancing the quality and efficiency of feature fusion. A joint loss function is constructed to comprehensively optimize the performance of the generator. Comparative tests on different datasets demonstrate significant improvements with the proposed algorithm. Furthermore， the super-resolved images exhibit higher performance in downstream tasks such as object detection， confirming the effectiveness and application potential of the algorithm in space-borne infrared remote sensing image super-resolution.

Abstract:Aiming at the problem that infrared small target detection faces low contrast between the background and the target and insufficient noise suppression ability under the complex cloud background， an infrared small target detection method based on the tensor nuclear norm and direction residual weighting was proposed. Based on converting the infrared image into an infrared patch tensor model， from the perspective of the low-rank nature of the background tensor， and taking advantage of the difference in contrast between the background and the target in different directions， we designed a double-neighborhood local contrast based on direction residual weighting method （DNLCDRW） combined with the partial sum of tensor nuclear norm （PSTNN） to achieve effective background suppression and recovery of infrared small targets. Experiments show that the algorithm is effective in suppressing the background and improving the detection ability of the target.

Abstract:The fusion of infrared and visible images should emphasize the salient targets in the infrared image while preserving the textural details of the visible images. To meet these requirements， an autoencoder-based method for infrared and visible image fusion is proposed. The encoder designed according to the optimization objective consists of a base encoder and a detail encoder， which is used to extract low-frequency and high-frequency information from the image. This extraction may lead to some information not being captured， so a compensation encoder is proposed to supplement the missing information. Multi-scale decomposition is also employed to extract image features more comprehensively. The decoder combines low-frequency， high-frequency and supplementary information to obtain multi-scale features. Subsequently， the attention strategy and fusion module are introduced to perform multi-scale fusion for image reconstruction. Experimental results on three datasets show that the fused images generated by this network effectively retain salient targets while being more consistent with human visual perception.

Abstract:Small target detection has been a classic research topic in the field of infrared image processing， and the objects are usually brighter than the local background. However， in some scenarios， the target brightness may be lower than the background brightness. For example， the civil airplanes usually have low-temperature skin when cruising， appearing as dark points on medium spatial resolution thermal infrared satellite images. There are few features of these objects， so the current detection networks are redundant. Hence， we proposed a lightweight dark object detection network， AirFormer. It only has 37.1 K parameters and 46.2 M floating-point operations on a 256×256 image. Considering the lack of infrared dark object detection dataset， the authors analyzed the characteristics of airplanes on thermal infrared satellite images， and then developed a simple simulation method for medium spatial resolution thermal infrared satellite images of civil aviation aircrafta， and constructed an infrared image weak target detection dataset IRAir using civil aviation aircraft as the simulation object. AirFormer achieves 71.0% at recall and 82.6% at detection precision on the IRAir dataset. In addition， after training on simulated data， AirFormer has achieved detection of real flying airplanes on the thermal infrared satellite images.

Abstract:Low-rank and sparse decomposition method （LRSD） has been widely concerned in the field of infrared small target detection because of its good detection performance. However， existing LRSD-based methods still face the problems of low detection performance and slow detection speed in complex scenes. Although existing low-rank Tuck decomposition methods have achieved satisfactory detection performance in complex scenes， they need to define ranks in advance according to experience， and estimating the ranks too large or too small will lead to missed detection or false alarms. Meanwhile， the size of rank is different in different scenes. This means that they are not suitable for real-world scenes. To solve this problem， this paper uses non-convex rank approach norm to constrain latent factors of low-rank Tucker decomposition， which avoids setting ranks in advance according to experience and improves the robustness of the algorithm in different scenes. Meanwhile， a symmetric GaussSeidel （sGS） based alternating direction method of multipliers algorithm （sGSADMM） is designed to solve the proposed method. Different from ADMM， the sGSADMM algorithm can use more structural information to obtain higher accuracy. Extensive experiment results show that the proposed method is superior to the other advanced algorithms in detection performance and background suppression.