2020 Vol. 49, No. 5
2020, 49(5): 20190454.
doi: 10.3788/IRLA20190454
Focal plane assembly of space remote sensor is a key component of space remote sensor, which is used to finish the transition function of the light signal into electrical signal. The focal plane assembly should have the characteristics of high splicing accuracy, high stability and high lightweight performance. In view of the function requirement of a deep space remote sensor with push-scan imaging and staring imaging, a focal plane assembly whose substrate was common with CCD detector and CMOS detector was designed, the focal plane assembly included focal plane substrate, CCD component, CMOS component, video processing circuit board, and so on. With a reasonable layout, the focal plane assembly was designed with three pieces of CCD and two pieces of CMOS, which was a compact structure and had a good heat dissipation effect. Moreover, the electromagnetic compatibility performance of focal plane assembly was improved obviously with the means of copper segment and box structure covering. After the assembling of focal plane assembly, the lapping error of CCD detector was better than ±2 μm, the collinearity and parallelism error of detector of CCD and CMOS was better than ±2 μm, the coplanarity error was better than ±2.5 μm. Finally, the force and thermal tests were carried out for the focal plane assembly, and the collinearity, parallelism and coplanarity errors of the detector did not change, indicating that the focal plane component has high enough splicage accuracy and stability. The fundamental frequency of the focal plane assembly is 135 Hz, which had sufficient dynamic stiffness. At the same time, the electromagnetic compatibility test was carried out on the space remote sensor, with good performance.
2020, 49(5): 20190458.
doi: 10.3788/IRLA20190458
Demand for autonomous navigation for mars exploration during the surround segment, a high-precision autonomous navigation method using Mars ontology was proposed based on the initial orbital parameters and the Mars image was taken by the optical navigation camera. The method first used the initial orbital parameters to obtain a projection of Mars in the camera phase plane, then matched the projection map with the Mars image to get the edge map of Mars, using the edge map for accurate ellipse fitting to achieve detector-Mars line of sight distance and line of sight vector measurement. In the process of algorithm simulation, the problems of image rotation, translation and size transformation that may be encountered in engineering implementation were fully considered. The simulation results show that the proposed image fusion measurement algorithm using prior information, the accuracy and reliability is improved more than the navigation method using mars image alone, which can meet the mission requirements of Mars exploration during the surround segment.
2020, 49(5): 20190460.
doi: 10.3788/IRLA20190460
Without atmosphere, the lunar surface is gradually evolved into lunar soil by the space weathering. The space weathering generates lunar soil and matures it. This progress changes the morphological structure and the spectral characteristics. Thus the spectral characteristics of lunar soil contains information about the space weathering of the moon's surface. Lunar soil maturity is an important indicator to describe the degree of the space weathering. Using hyperspectral data to produce the submicroscopic metallic iron (SMFe) and to obtain lunar maturity is an important method for studying the space weathering. Since the in situ detection data is not disturbed, the results obtained are relatively more accurate and reliable. Therefore, the Chang'e-4 (CE-4) Yutu-2 rover was equipped with a detection instrument that directly obtained the in situ hyperspectral data (450-2395 nm) of the lunar surface, which provided a good opportunity for studying the lunar weathering. Two spectral data near the landing site of the CE-4 lander were selected, using the Hapke model and the spectral angle matching method, SMFe content was produced. The maturity of the lunar soil near the landing site of the CE-4 was further inferred based on the Morris model and FeO content. The results show that the lunar soil's maturity is 11.5, indicating that it may be immature lunar soil.
2020, 49(5): 20190463.
doi: 10.3788/IRLA20190463
The dual resolution camera combines large field of view and high resolution. The coaxial optical fixed structure avoids many problems caused by the moving parts of the zoom lens, and has potential application value in deep space detection target tracking and smart terminals such as mobile phones. The existing dual resolution image zoom algorithm based on deep learning has slow speed, no increase in information, poor adaptability of image network structure and forgery of image information restoration. To join a solution in normal image algorithm based on depth information was tentatively proposed in this paper. The feasibility of using image focus sharpness as depth information into the dual resolution zoom algorithm was demonstrated, the accuracy and effect of focus depth information detection was explored by Laplace image evaluation, and the dual resolution zoom algorithm of deep learning and traditional methods based on depth information was tested. The new excellent zoom algorithm used the space during normal focus of the camera which did not affect the imaging speed of normal images. The algorithm complexity is reduced by 60%, the computational memory overhead is reduced by 35%. The spatial relationship super-resolution information is more realistic and reliable, and the image result evaluation is improved by 10% to 50%.
2020, 49(5): 20190467.
doi: 10.3788/IRLA20190467
In Lunar Laser Ranging (LLR), higher repetition rate is an important trend for development, which is also of great significance for acquiring more LLR data. Based on the previous 10 Hz ranging rate system, a 100 Hz lunar ranging system with common-optical-path structure was developed by Yunnan Observatories, CAS. However, overlapping of the transmitting/echo signals exists in such a common-optical-path system, and seriously disturbs echo-photon detection. In the paper, the overlapping phenomenon was analyzed from the perspective of time sequence, and the correlation between the overlapping and the Time of Flight (TOF) was studied. Overlapping in the system was simulated with the orbital prediction of a lunar retro-reflector of a certain day, and the general overlapping ratio was approximately 8% for the observational duration. Sectional simulation based on each transmitting epoch was also performed, its result was consistent with the previous one. The method of avoiding overlapping by slightly adjusting rotating-mirror's speed was analyzed, and corresponding experiment was carried out, as proper speed adjustment turned out to be a solution when the overlapping happened.
2020, 49(5): 20190471.
doi: 10.3788/IRLA20190471
In the deep space exploration, the signal is delayed by the troposphere and a certain delay will occur at the receiver end, which affects the detection accuracy. The existing methods mainly realize the delay prediction by the grid model and the space model, but the accuracy of the model is limited due to the regional difference, and the prediction accuracy still has room for improvement. In this paper, a deep-space tropospheric delay prediction model based on adaptive multiple-input multiple-output (MIMO) signals was proposed. The satellite signal MIMO transmission mode was simulated based on a single transceiver antenna, and then an adaptive Kalman filter was constructed. The optimal transmission path was selected by adaptively adjusting the weight coefficient of the MIMO signal component to predict the tropospheric delay. The number of satellites participating in the measurement was four. Experiments were carried out under different signal-to-noise ratios and changing the number of MIMO channels to study the accuracy and actual measurement error of the adaptive MIMO model. The experimental results show that the prediction accuracy of the new method is much higher than that of the GPT2 model, GPT2w model and the commonly used UNB3 model and EGNOS model in real-time navigation and positioning.
2020, 49(5): 20190555.
doi: 10.3788/IRLA20190555
In order to analyze the reasons for star sensor performance degradation and the decrease of attitude measurement accuracy, performance degradation mechanisms of star sensor caused by Complementary Metal Oxide Semiconductor (CMOS) image sensor radiation damage were mainly studied. By establishing the correlation among the spatial radiation, radiation sensitive parameters of CMOS image sensor and the performance parameters of star sensor, the transmission mechanism from CMOS image sensor device parameters degradation to star sensor system parameters degradation was revealed. The 60Co-γ irradiation test showed that the decrease of signal-to-noise ratio led to the decrease of star sensor detection sensitivity after irradiation, and the signal-to-noise ratio was the bridge between the CMOS image sensor and the star sensor system. The proton irradiation test showed that when the irradiation fluence was more than 3.68×1010 p/cm2, the star point centroid could not be extracted correctly. The results lay a certain foundation for the research of star sensor attitude measurement error and correction technology, and also provide some theoretical basis for the design of high-precision star sensor.
2020, 49(5): 20201001.
doi: 10.3788/IRLA20201001
Deep space exploration refers to the exploration of the moon and the celestial bodies or space environment beyond the moon. The field of deep space exploration is an important part in implement strategy of space power, and is also one of the signature field of the spaceflight. This paper summarizes the development of deep space explorations in the past 60 years, highlights the achievements and future development trends of deep space exploration in China. According to Chinese Lunar Exploration Program(CLEP) and the first Mars exploration missions, the key technical breakthrough in these task were analyzed. Breakthrough, focusing on the follow-up mission planning of lunar and deep space exploration, discussed the key technologies of missions such as the establishment of lunar research station, asteroid exploration, Mars sampling return and Jupiter system exploration. On this basis, it summarizes the development route of China's deep space exploration technology and the key technologies that need to be studied and overcome in the future.
2020, 49(5): 20201002.
doi: 10.3788/IRLA20201002
Optical technology is widely applied in the China Lunar Exploration Program (CLEP) and the first China Martian exploration project: More than half of the scientific payloads were optical remote-sensing payload, and it was also widely used in the engineering areas including autonomous navigation, TT&C, monitoring, rendezvous and docking. In this paper, accomplished CLEP missions, on-going Chinese lunar and Martian projects and especially main achievements of optical technology applications were overviewed. Roadmap on China's future lunar, mars, jupiter and asteroids exploration was introduced. Development trends and prospects of optical technology in lunar and deep space exploration were analyzed and proposed.
2020, 49(5): 20201003.
doi: 10.3788/IRLA20201003
Deep space optical communication is the main technical approach of high speed data transmission to meet the future deep space exploration, and the main technical direction for the China Deep Space TT&C Network development. Especially with the development of intelligent and networked space communication system, the intelligent space communication network combining RF and optical communication will become the core facilities to support future deep space exploration missions. This paper combines China future lunar and deep space exploration development plan, based on the mission support capability of China Deep Space TT&C Network, the requirements of future deep space optical communication are sorted out. On the basis of learning from the latest research results, this paper analyzes the technical approach of constructing the earth terminal of deep space optical communication system. Based on the actual situation of China, this paper puts forward the development idea of constructing the mixed deep space optical communication ground system with space relay terminal and earth terminal, and finally forming the cognitive network of deep space optical communication.
2020, 49(5): 20201004.
doi: 10.3788/IRLA20201004
Optical sensor is the key component of the navigation guidance and control system in the deep space exploration mission, and it provides direction and location for stable operation of the spacecraft and scientific detection of the payload. The typical products of attitude sensor, autonomous navigation optical sensor, landing and obstacle avoidance optical sensor, patrol detection optical sensor were reviewed, and the key technology and development tendency were discussed.
2020, 49(5): 20201005.
doi: 10.3788.IRLA20201005
Spectral imaging technology is an important technical means in the field of deep space exploration and can serve a variety of space scientific experiments. Aiming at the demand of high stability and high light utilization of spectral loads in the field of deep space exploration, a new deep space hyperspectral diffraction computational imaging detection technology was proposed . Combining diffraction spectrum imaging technology and light field imaging technology to solve the problem of large-scale scanning of the detector in the diffraction spectrum push-broom imaging system, and one shot was realized to obtain the target spatial information and spectral information, that is, snapshot imaging. This has guiding significance for the design and research of deep space exploration hyperspectral imagers.
2020, 49(5): 20201006.
doi: 10.3788/IRLA20201006
Exploring the chemical composition of materials on the surface of the moon and planets is the key to research its origin and evolution history. Spectral detection technology is the major methods of material composition recognition and quantitative inversion. In-Situ spectral detection is different from remote sensing and sampling return detection, it refers to the short distance spectral detection performed at the target site. The lunar science research and resource exploration of "Chang'e-3 "mission need to implement the first In-Situ spectral detection technology on the lunar surface, by breaking through the key technologies of the new acousto-optic spectroscopic detection of staring time series scanning, we have developed high-performance, light and small, highly reliable instruments, adapted to the harsh surface environment, and become the first in the world to realize In-Situ spectral detection and analysis on the lunar surface of the moon. The paper takes "Chang'e-3 and 4" as typical applications, and introduces the In-Situ spectral detection technology of the lunar surface, including detection mechanism, working mode and the function, performance and application of the instrument; Finally, the "Chang'e-5" lunar mineral spectrometer was introduced which will be applied soon.
2020, 49(5): 20201007.
doi: 10.3788/IRLA20201007
The pace of human exploration of the universe has never stopped. To go further is the unremitting pursuit. As early as 1969, the ''Apollo ''11 spacecraft of the United States realized the first landing of human beings on the moon, and then another five spacecraft successfully landed on the moon. In recent years, China's "Chang'e" project as a representative of the successful implementation of the unmanned lunar exploration program, making the lunar exploration once again by the world's attention. In the context of the continuous development of space exploration, manned moon landing will be on the stage again at some time in the future. The important processes of landing on the moon, lunar surface activity, take-off and ascent are all inseparable from the control technology. After introducing the early manned landing on the moon and the achievements of lunar exploration in recent years, this paper analyze the concept of the future manned landing mission in China, discusses the crucial control technology, and puts forward several development prospects.
2020, 49(5): 20201008.
doi: 10.3788/IRLA20201008
In orbit capturing and real-time monitoring the status of distant dimmer space targets is of vital importance for space safety. However, when the sensitivity approaches 14 magnitude, the intensive stellar population, the stellar scattering light will reduce the detection efficiency greatly, resulting in a limited detection range and ability. To solve this problem, a simple method for space object capturing, classification and online tracking was presented, synchronously eliminating the dense background stars and sequentially extracting the possible space targets, utilizing the regular motion of a moving object in a short timespan. Ground simulations and tests validate the effectiveness and accuracy of the method. It seems that this method can also be used as a basic function for deep space projects, such as asteroid and planet exploration in the solar system.
2020, 49(5): 20201009.
doi: 10.3788/IRLA20201009
Autonomous optical navigation is a key technology for asteroid exploration missions. Terrain relative navigation methods based on optical images have been successfully applied in previous asteroid exploration missions. Optical navigation uses features such as the asteroid's surface topography and albedo to accurately establish reference datums and navigation landmarks, and give precise relative positions, attitudes, and velocities during descent based on the matching results of real-time observation images. The development of optical navigation technology for asteroid probe descent and landing was reviewed, and its key technologies and development trends were summarized, which provided a reference for China's asteroid exploration mission.
The cryogenic spectrometer for the airborne thermal infrared hyperspectral imager (ATHIS) was designed in this paper. In order to detect the spectral recognition capability of the ATHIS, detailed spectral characteristics tests were carried out in the laboratory. The method of CO2 laser combined with high precision monochromator was proposed to achieve the absolute spectral calibration accuracy of 0.8 nm. Methane gas was tested in the laboratory, and the results show that methane gas could be accurately identified. Twice flight tests of the ATHIS was carried out based on those experiments. The data processing results show that thermal infrared hyperspectral imager can effectively carry out classification and identification of typical urban buildings and effective monitoring of industrial chemical gas emissions. The above preliminary application shows that the ATHIS has excellent spectrum recognition power, and the primary business application prospect. The follow-up instrument will carry out the researches on the enhancement of radiation quantification.
2020, 49(5): 20190361.
doi: 10.3788/IRLA20190361
Infrared low-background detection technology is mainly used in space environment, the sensitivity of detection system has a gret relationship with the background radiation of detection system. How to effectively suppress the background radiation of detection system has always been the important research direction. And recognition has been a research hotspot in infrared field, adding the detection spectrum is the most effective way to get the features of target. An infrared multi-band detection technology based on optical lens local refrigeration and optical spiltter was introduced. Firstly, adopting the idea of integrated design with optical system and detector, integrate local optical system into the infrared detector and refrigeration with infrared FPA, then dividing the two lights by optical spiltter, which achieved dual-band ability, if the dual-band FPA was used in the future, the system can have the multi-band ability. Comparing the results of self-radiation under different optical types by the simulation of optical self-radiation, it can be seen that the background radiation of this technology is reduced to 1/4 of the ordinary temperature optical system, the sensitivity has also been greatly improved. This technology has obvious advantages and great potential.
2020, 49(5): 20190446.
doi: 10.3788/IRLA20190446
To meet the requirements of intrusion detection with high recognition rate and low false alarm rate in specific areas such as airports and oil depots, a target recognition method based on optical fiber sensing and infrared video was proposed. Among them, the distributed optical fiber vibration sensor based on MCSVM ADMZI (asymmetric dual Mach-Zehnder interferometer) was used in the optical fiber sensing part, which combined the empirical mode decomposition (EMD) and the kurtosis feature with the MCSVM to improve the recognition rate. The infrared recognition part improved the clarity of the gray difference image through the wavelet transform. The intrusion detection was realized by pattern comparison algorithm. The field experiment results show that this method can identify four common intrusion events (climbing fence, tapping cable, cutting fence, shaking fence). The average recognition rate is over 92.5%, and the false alarm rate is 0.9%. Compared with the traditional single sensor scheme, this method has a great improvement in the system performance such as false alarm rate and false alarm rate, and can meet the practical application requirements.
2020, 49(5): 20190457.
doi: 10.3788/IRLA20190457
The space environment of the three-axis stabilized satellite in geosynchronous orbit is complex, and the background radiation of the instrument varies greatly. The traditional stray light analysis method can not simulate the non-uniform temperature field, and can not calculate the instrument background in real time. Also the simulation error is large. Thermal radiation stray light integration method was used to analyze instrument background radiation of in-orbit infrared camera. The background radiation of the instrument and the signal-to-clutter ratio of the camera on the detector was calculated by using the real-time temperature fields with temperature gradient and radiation transfer factor. Comparing the result of the thermal radiation stray light integration method and traditional stray light analysis method with the on-orbit measurement, the error of the thermal radiation stray light integration method was less than 17%, while the error of traditional stray light analysis method was up to 114%. The result shows that the thermal radiation stray light integration method is closer to the actual on-orbit situation, and the simulation efficiency and accuracy are higher.
2020, 49(5): 20190462.
doi: 10.3788/IRLA20190462
In order to improve the stability and resolution of equivalent temperature and area, an infrared multi-spectral design method was proposed based on feature parameter extraction of point target. Firstly, the anti-error ability of multi bands was simulated by multi spectral design criteria, and the influencing factors of spectrum design were determined. According to the noise analysis of the ground-based infrared radiation measurement system, the conditions for the best working performance of the middle wave and long wave photodetectors were determined. Finally, the spectral distribution was determined according to the estimation of the band SNR and the infrared atmospheric window. The performance analysis results show that the designed spectrum can make the medium wave and the long wave work together in the case of various changes of the target, greatly improving the stability and resolution of the equivalent temperature and the equivalent area.
2020, 49(5): 20190491.
doi: 10.3788/IRLA20190491
A novel ultra- low-power readout integrated circuit (ROIC) for 1 024×1 024 ultraviolet (UV) AlGaN focal plane arrays (FPA) with 18 μm-pitch was presented. In order to optimize power consumption for UVFPA readout circuit these methods were adopted, which including single-terminal amplifier under subthreshold region as CTIA amplifier, common current source load for source follow (SF) buffer in column pixels and level shift circuits, and time-sharing tail current source for column buffer. The smallest operational current of CTIA in pixel unit is only 8.5 nA with 3.3 V power supply by using single-terminal amplifier. The ROIC has been fabricated in SMIC 0.18 μm 1P6M mixed signal process and also achieved better performances with the novel design of bias current adjustable. Furthermore, the overall power consumption of the chip is 67.3 mW at 2 MHz in 8-outputs mode by the above methods according to the experimental results.
2020, 49(5): 20190493.
doi: 10.3788/IRLA20190493
The high temperature gas such as H2O and CO2 of the exhaust plume of a launching ballistic missile will bring strong IR radiation, so it is the aero heated warhead and the high temperature gas flow around the warhead. The IR radiation is an important signal for IR early warning, tracking and guiding of target in the near-space. The IR radiation characteristics of Angi-Ⅱmissile during launching and reentry stage was studied. Based on narrow band radiation model, and the radiation mechanism of vital gas species were taken into account. The methods for spectral parameters calculation of the species in the high temperature flow were set up, and then a computer code was developed for IR characteristics of targets in the air, which can compute IR characteristics of warhead body and high temperature flow. According to simulated data of the exhaust plume flow and reentry flow around the warhead, the radiation code was used for computing the IR characteristics of Angi-Ⅱ at launching and reentry stage, which can be used as references for early warning and anti-ballistic missile about Angi-Ⅱ.
2020, 49(5): 20190413.
doi: 10.3788/IRLA20190413
Infrared radiometry technology is an important means to characterize the infrared signature of targets, and atmospheric correction is a requisite step to obtain the real radiance of targets. A nonlinear atmospheric correction (NLAC) method was proposed to improve the infrared radiometric accuracy for long distance targets in this paper. This method used near-range standard reference source measurement (NRSRM) to calculate the actual atmospheric transmittance and path radiation simultaneously at different locations in a real-time environment. And the theoretical atmospheric transmission and path radiation under the corresponding conditions could be obtained from the atmospheric radiation transmission software as well. Neural network technology was applied to fit the non-linear relationship between them. Thus, the atmospheric transmittance and path radiation over long distances could be predicted to achieve atmospheric correction. Simpler linear atmospheric correction (LAC) and linear enhancement atmospheric correction (LEAC) were also carried out for comparison. The experimental results indicate that the infrared radiometric average error of the proposed method is 6.45%, which is much lower than that of the conventional method, LAC and LEAC that are 16.17%, 11.27% and 7.44%, respectively.
2020, 49(5): 20190552.
doi: 10.3788/IRLA20190552
Two-person interaction recognition based on multi-stream spatio-temporal fusion was proposed. Firstly, a method to describe two-person’s skeleton which invariable with angle of view was proposed. Then a two-layer spatio-temporal fusion network model was designed. In the first layer, the spatial correlation features were obtained based on one-dimensional convolutional neural network (1DCNN) and bi-directional long short term memory(BiLSTM). In the second layer, the spatio-temporal fusion features were obtained based on LSTM. Finally, the multi-stream spatio-temporal fusion network was used to obtain the multi-stream fusion features, which learned one kind of feature by one stream and fusion features for all streams together at last. The weights for each stream was shared, and every stream had the same structure. After features were fusion for all streams, it could be used for interaction recognition. By applying this algorithm to NTU-rgbd datasets, the accuracy for two person interaction recognition for cross-subject could reach 96.42%, and the accuracy of two person interaction recognition for cross-view could reach 97.46%. Compared with the state of art methods in this field, this method performed best in two person interaction recognition.
2020, 49(5): 20190553.
doi: 10.3788/IRLA20190553
Vision chip is a high-speed, low-power intelligent vision processing system on a chip (SoC), it can be applied to many different fields of production and daily life. A new programmable vision chip architecture was proposed which took into account the computational features of traditional computer vision algorithms and convolutional neural networks, it enabled the architecture to support the two types of algorithms simultaneously. The vision chip integrated multi-level parallel programmable processing array, high-speed data transmission path and system control module, it was fabricated in a 65 nm standard CMOS process. The experimental results show that the vision chip achieves 413GOPS peak performance at 200 MHz system clock and can perform various computer vision and artificial intelligence algorithms including face recognition and target detection efficiently. The proposed work shows higher system performance in terms of programmability, performance and energy efficiency when compared with other state-of-the-art vision chips.
2020, 49(5): 20200015.
doi: 10.3788/IRLA20200015
For underwater image of low contrast, color deviation and blurred details and other issues, the multi-input fusion adversarial networks was proposed to enhance underwater images. The main feature of this method was that the generative network used encoding and decoding structure, filtering noise through convolution layer, recovering lost details through deconvolution layer and refining the image pixel by pixel. Firstly, the original image was preprocessed to obtain two types of images: color correction and contrast enhancement. Secondly, the confidence graph of the difference between the two enhanced images and the original image was learned by using the generated network. Then, in order to reduce artifacts and details blur introduced by the two enhancement algorithms in the process of generating network learning, the texture extraction unit was added to extract texture features from the two enhanced images, and the extracted texture features were fused with the corresponding confidence map. Finally, the enhanced underwater image was obtained by constructing multiple loss functions and training the adversarial network repeatedly. The experimental results show that the enhanced underwater image has bright color and improved contrast, the average value of UCIQE and NIQE is 0.639 9 and 3.727 3 respectively. Compared with other algorithms, the algorithm has significant advantages and proves its good effect.
2020, 49(5): 20201010.
doi: 10.3788/IRLA20201010
In order to solve the problem that the Capsule network increases the amount of calculation and the number of parameters increases sharply with the input picture, the Capsule network is improved and the improved Capsule network is used in SAR automatic target recognition (SAR-ATR). In this paper, based on the mechanism of brain visual cortex processing information in hierarchical structure and column form, the idea of complete instantiation was proposed, and the brain-like calculation was used to improve the Capsule network. The specific method was to use multiple convolution layers to achieve hierarchical processing. The number of convolution kernels used in each layer increases with the depth of the hierarchy, which made the extracted abstract features gradually increase. In the PrimaryCaps layer, the Capsule vector consisted of all the feature maps output by the last layer of the convolutional layer, so that the Capsule unit contained all the features of the target part or the whole to achieve full instantiation of the target. On the SAR-ATR, a comparison experiment was performed with the Capsule network, the traditional target recognition algorithm and the target recognition algorithm based on the classical convolutional neural network. The experimental results show that the improved Capsule network training parameters and calculations are greatly reduced, and the training speed is greatly improved, and the recognition accuracy on the SAR image data set is increased by 0.37 and 1.96-8.96 percentage points compared with the Capsule network and the first two methods respectively.
2020, 49(5): 20190290.
doi: 10.3788/IRLA20190290
In recent years, with the development of heterodyne array receivers of radio frequency (RF) telescope, beam splitters based on phase grating technology have gained important applications in the sub-millimeter wavelength range. It is capable of transmitting a single local oscillator signal to the superconducting SIS/HEB mixer receiver array by splitting beam synchronization. Due to the characteristic size of the terahertz phase grating at sub-micron level, its machining precision directly affects the performance of the device, which brings great challenges to the micro-machining technology. Based on this, a new terahertz four beam splitter combined with phase coding metamaterials was designed. Only one single-layer metamaterial coding unit can be used to realize the separation of broadband electromagnetic beam and high beam conversion efficiency. The structure of THz beam splitter is simple and easy to process, while the direction of the reflected beam can be flexibly adjusted. In order to match the actual test system, we focused on beam separation at different angles of incidence and obtain the best range of oblique incidence angles (less than 30°). The relative bandwidth is 52%, and the reflected four beam powers differ by no more than 10%. This provides a new solution for the local oscillator signal power distribution of terahertz superconducting mixer array receivers of RF telescope, and also facilitates the design and development of other new terahertz functional devices.
2020, 49(5): 20190338.
doi: 10.3788/IRLA20190338
In view of the large diameter and different transmission channels of terahertz beam, the large area terahertz pyroelectric detector and the multi-purpose detector structure were proposed to measure the terahertz beam power, which was transmitted in free space and waveguide. Firstly, a terahertz pyroelectric detector model was established by using the finite element analysis software, and the thermoelectric coupling simulation design was carried out. Secondly, the terahertz pyroelectric detector was developed by using precision grinding and polishing process, micro-nano processing technology of planar integrated circuit, homogenized and peeling process, and grinding wheel slicing technology. Finally, the sleeve and waveguide adaptor mounted on the detector structure were innovatively designed. Theoretical analysis and experimental results show that the terahertz pyroelectric detector designed in this method has the characteristics of low noise equivalent power and high repeatability, and solves the compatibility test problem of terahertz radiation power transmitted in free space and waveguide.
2020, 49(5): 20190489.
doi: 10.3788/IRLA20190489
Metamaterial perfect absorber is a typical functional electromagnetic material and holds great potential in application fields such as highly efficient ulitization of solar energy. Up to date, most works are dovoted to the tunability of working wavelength as well as realization of dual-, triple- and even broad-band perfect absorption. However, applications such as laser protection require that the metamaterial peferct absorber can function at specific wavelength with narrow bandwidth, and work devoted to this direction is rather rare. In this work, based on a triple-layer structure made of Al back mirror–SiO2 spacer–Al disk, a narrow-band metamaterial perfect absorber working at 1 064 nm was proposed and studied by numerical simulations. It was found that, compared to utilizing the fundamental mode of plasmonic resonance of smaller structure unit, utilization of higher-order mode of larger structure unit allowed for a narrower bandwidth of perfect absorption at designated wavelength. In addition, through systematic study of the geometric parameters such as the thickness of spacer layer, the diameter of the Al disk and the period of the lattice, the influence of such geometric paremeters on the optical response of metamaterial perfect absorber was revealed. Based on these findings and through optimization of geometric parameters, a high performance and narrow-band metamaterial perfect absorber design was obtained with a zero transmittance, a reflectance as low as 8.56×10−5 and a bandwidth of about 55 nm. Since all the materials involved are CMOS compatible, and the characteristic dimensions of the structure unit lie within the range of feasible fabrication method such as optical lithography, the metamateiral perfect absorber proposed here holds good promise towards large-scale real-world applications.
