This PDF file contains the front matter associated with SPIE Proceedings Volume 11563, including the Title Page, Copyright Information, and Table of Contents.

Highly uniform ball-shaped indium bump arrays with small pixel pitches down to 10 μm have been fabricated. Multilayer stacked metal contact electrodes covered by a thin SiN_x dielectric layer serve as the under bump metallization. Indium bumps were thermally evaporated inside the SiN_x openings on top of the electrodes. Wet lift-off of the indium bumps was achieved by using a negative photoresist with precisely controlled undercuts. By comparison to a recipe without the SiN_x, the non-uniform reflow effect of the indium materials was effectively eliminated after the thermal treatment. A mean indium ball diameter of 6.05 μm with a small coefficient of variation of 2.6% was finally realized for 10 μm pitch arrays. These results demonstrate the fabrication method is promising to ensure a reliable flip-chip hybridization of ultra-fine pitch focal plane arrays to silicon readout circuits with high yield.

Salient object detection (SOD) has become an active research direction with extensive applications in computer vision tasks. Although integrating RGB and infrared thermal (RGB-T) data has proven to be effective in adverse environments, it is difficult for RGB-T SOD methods to highlight the salient objects completely when objects cross the image boundary. To address the aforementioned problem, this paper proposes an effective RGB-T SOD algorithm based on multi-spectral co-connectivity (MSCC) and collaborative graph ranking. Specifically, we introduce the multi-spectral weighted color distance to construct an improved undirected weighted graph and compute the MSCC-based saliency map. Simultaneously, the MSCC-based background probability map is also calculated and employed in the following processing of real background seeds selection. Then, we utilize collaborative graph learning (CGL) and calculate the CGL-based saliency map in a two-stage ranking framework. Finally, we integrate these two saliency maps through multiplying or averaging to enhance the final saliency result. The experimental comparison results of 5 quantitative evaluation indicators between the proposed algorithm and 9 state-of-the-art methods on RGB-thermal datasets VT821 and VT1000 datasets demonstrate the robustness and superiority of the proposed work.

Regarding the radiation relationship between complex objects involved in the space environment, in this paper we use discrete coordinate method to calculate and analyze the coupled radiation-conduction heat transfer between complex objects. We simulate the heat conduction process by use of the finite volume method, and the overall temperature field is finally obtained. By comparing the calculation results of the program and Fluent software, the correctness of the coupled radiation-conduction heat transfer program is verified. The result shows that the radiation heat transfer between the targets could affect the temperature field under the conditions of self-emission and ambient radiation. And it also shows that an appropriate reduction of discrete points can reduce the calculation time and will not affect the calculation accuracy.

To meet the growing need for calibrated detector at long-infrared wavelengths, we are reporting on a laser-based absolute spectral responsivity scale of the detector in the long-infrared spectral region. The high-accuracy cryogenic radiometer (HACR) is a cryogenic electrical substitution radiometer that serves as a primary standard for optical power measurements. We measured the transfer standard HgCdTe/sphere detectors at 9.28μm, 9.62μm, 10.26μm and 10.60μm, respectively. The experimental results show that the extended uncertainty of optical power measurement was less than 0.30%～0.42% (k = 2); The extended uncertainty of the absolute spectral responsivity measurement has been analyzed to be 0.80%～1.02% (k = 2).

The performance of InP/InGaAs SPAD detectors depends on the electrical field distribution in their multiple layers. In the conventional separate absorption, grading, charge and multiplication (SAGCM) structure, the major function of the charge layer is to confine the electrical fields, so the charge layer’s parameter design is very important for any enhanced SPAD detectors. Normally the sheet density equals to doping concentration times thickness is considered as one of the key factors for the device design, however, even with the same sheet density, there are different combinations of doping densities and thicknesses. Our calculations show that with the same sheet density of charge layer, the one with higher doping concentration has higher electrical fields in both multiplication and absorption layers, then has lower breakdown and punch-through voltages. The results were also verified by the experimental measurements.

Innovations in optics, focal plane arrays (FPA), microelectronics and image processing have revolutionized infrared camera design. Infrared detectors coupled with nanostructures are expected to subvert the traditional intensity detectors and advance to the multi-dimensional and data cube detectors. Photon trapping technology is expected to decrease the intrinsic dark current, broaden the response band, increase the QE and increase the working temperature. High density FPAs result to novel wide field-of-view small SWaP broadband infrared camera architecture. Digital-pixel focal plane for real-time digital image processing enable the intelligent chips. This paper highlights sensors with nanostructure coupled FPA, photo trapping FPAs, high density FPAs and digital-pixel FPAs.

Indium bumps are widely employed in high density interconnection between infrared focal plane arrays and Si read out integrated circuits by flip-chip bonding. Indium bump array formation is a critical step in the flip-chip fabrication process. Taller and higher uniformity indium bumps are necessary for high pixel density and low noise photodetectors. In this work, a new process of indium bumping through evaporation and ion etching was developed to produce ultrafine pitch indium bumps for assembly of large-area HgCdTe photodetector. Electron microscopy was used to analyze and evaluate the microstructure, height and uniformity of indium bumps. The results showed 7 μm height indium bumps with 10 μm pitch were easily achieved. The bump height and uniformity were significantly improved with our new developed indium bump fabrication method.

At present, HgCdTe arrays has been commonly used in photodetectors for infrared detection in space system, where they are exposed to the space radiation environment. For high radiation-tolerance and the optimum performance of the detector, surface-passivation technology that provides long-term stability is required. Double layer of ZnS/CdTe passivant is most recommendable, because ZnS/CdTe-passivated HgCdTe detectors could show great insulating and radiation-tolerant properties. However, the thickness ratio of ZnS and CdTe layers has not been optimized. In this study, the gamma radiation effects on ZnS/CdTe-passivated mid-wavelength HgCdTe arrays with different ZnS-layer thicknesses were investigated, by analyzing the current-voltage curves before and after gamma irradiation at a very low rate. To our surprise, increase of ZnS thickness from 300 nm to 700 nm dramatically improved the radiation tolerance, although ZnS is considered to be vulnerable to gamma irradiation. We hypothesized that gamma radiation could be strongly absorbed by ZnS with sufficient thickness and the transmitted intensity is harmless to the HgCdTe arrays. Therefore, increase of ZnS thickness could protect the HgCdTe arrays from gamma radiation damage. Here, we presented an efficient and easy processing method to increase radiation-tolerant properties of the high performance HgCdTe photodetectors.

The solar stray light will seriously affect the imaging quality of the infrared detection system. The traditional point source transmittance (PSF) index is difficult to evaluate the image degradation of the detection system due to stray light. In this paper, a stray light simulation is performed firstly for a given infrared detection system, and the irradiance distribution of solar stray light on the surface of the detector with different incident angles are obtained. Then a simulation model of the detection system is established based on the physical conversion process of the infrared detection system. By combining this simulation model with the light distribution, the spatial noise of the detection system with solar stray light can be calculated quantitatively, and the image degradation due to solar stray light can be quantitatively evaluated.

Mercury cadmium telluride (HgCdTe) grown by liquid phase epitaxy (LPE) demonstrates superior performance in the infrared imaging applications. The HgCdTe devices are fabricated by depositing an epitaxial layer of HgCdTe on cadmium zinc telluride (CdZnTe or CZT) substrates via LPE. This LPE growth requires high-quality substrates. However, it is difficult to manufacture and polish epi-ready substrates for the LPE growth of II-VI semiconductors, especially for HgCdTe. This leads to very high cost and limits development of systems based on LPE-grown HgCdTe films. For very large two-dimensional components, thermal expansion between HgCdTe and the silicon should be considered. For these components, the CdZnTe substrates of the detector chips are thinned by polishing and chemical etch technology, and removed totally to spread the stress between the HgCdTe layers and the Indium bumps. Therefore, the high-cost CdZnTe substrates could only be used once using the traditional fabrication process. In this paper, we present a reutilization process for CdZnTe substrate. Our results demonstrate the device prepared by HgCdTe film using repolished CdZnTe substrates has good property. We could produce more chips from one CdZnTe substrate with enough thickness. This method enable the reuse of the CdZnTe substrates and significantly decreases the cost of the device.

HgCdTe is significant material prepared infrared detector. Higher performance infrared detector and longer wavelength are vital requirements of the third infrared detector. Device performances of HgCdTe infrared detector and material interface state are inseparable. On the one hand, the HgCdTe lattice is zinc-blende cubic structure, when it is exposed to atmosphere, self-generating oxide layer can be forming on material interface, a good deal of fixed charges are existing in the oxide layer, that can make the device properties degradation seriously. On the other, in the traditional preparation process, epitaxial HgCdTe layer is corroded before surface passivation, material surface could form residual rich tellurium layer. It can also make the device properties degradation on account of recombination centers, increase surface leakage of the device. Based on the above, study on the surface treatment process of HgCdTe detector is very necessary, this paper researched a new method wiping out the self-generating oxide layer and residual rich tellurium layer to obtain a good in function HgCdTe infrared detector. In this way, prepared HgCdTe infrared detector containing cutoff wavelength of 5 μm, 9μm and 12.5 μm, compared HgCdTe infrared detector employing this new method and unused this new method. The former possessed lower dark current, lower noise equivalent temperature difference (NETD) and higher yield. This new method applying to process of HgCdTe infrared detector enhanced the detector properties, lay the foundation of longer wavelength HgCdTe infrared detector.

With the development of infrared technology, large-array infrared focal plane detector is getting more and more popular. In order to solve the problem of reduced reliability when the cooling head is packaged with large-array infrared detector, a new design with reinforcement structures on cold platform and cold shield is introduced in this paper. The finite element simulation is used to analyze the optimized structure, and the results show that this design can fully meet the mechanical requirements of the project. By using the reinforcement structures, a Dewar was designed and processed to package a 2.7K×2.7K-15μm large-array infrared detector. After all the packaging process, the total mass of the Dewar is only 800g, and the cooling head is more than 90g. Under the circumstances of a heavy cooling head, the Dewar successfully passed the environmental adaptability assessment, and there is no problem with all the indexes of the infrared focal plane detector, which verified the reliability of the reinforcement structure.

Dark current is a critical index for Hg_1-xCd_xTe IRFPA(Infrared focal plane array) detectors, which is main limit for the application in the very long wave(≥12μm) spectral range. In this paper we reduced the dark current by using small pixel effective areas. And by fabricating in-situ integrated microlens on the illumination side of the detector, the disadvantage of the small pixel effective areas has been overcome.

In order to solve the problem of small data set, this paper uses the invariance of distinguishing features between the simulated infrared image of maritime ship and real infrared image of maritime ship, studies a method of detecting infrared maritime ship target with no real data. At the same time, we propose an attribute adaptive learning strategy based on deep learning algorithm of yolov3. In the case of low data support, the detection capabilities of infrared maritime ship target have been improved.

With the development of InSb infrared focal plane array (IRFPA) detectors, the scale of the detector array is growing larger, the size of the pixel and the space between pixels are becoming smaller. To realize the fabrication of high quality mesa-structure device, the paper focuses on the study of inductively coupled plasma ( ICP) dry etching technology using etching gas system based on methane and chlorine to realize mesa etching of InSb material. The paper compares the etching principle of two etching gas systems as CH₄/H₂/Ar and BCl₃/Ar. Etching results such as etching morphology, etching rate, etching surface stoichiometric ratio and I-V test are compared. Through the research, BCl₃/Ar gas system is found faster in etching rate. Etching with both kind of gas systems can result in smooth etching surface, normal stoichiometric ratio and qualified I-V test result which satisfy technological requirements.

The underfill technology of a large-scale infrared focal plane detector chip is studied. The flow characteristics and principle of filling glue were analyzed. By changing the chip angle, the gap between the detector chip and the readout circuit is filled by fluid capillary action of spontaneous siphon phenomenon, and the incomplete filling caused by the rapid flow of glue around the edge of the chip was avoided at the same time. The formula of filling time of vertical dispensing is given, and the relationship between the process form, design appearance of large-scale infrared focal plane detector chip and dispensing process was analyzed.

The infrared detection technology of the dim point target in complex background is of great significance to the development of infrared search and tracking (IRST) systems, precision guidance and other fields. The dim point target at the long distance occupies very few pixels in the infrared image, and there is no detail and texture feature. Moreover, the imaging process is always affected by the nonuniform noise, complex clouds, atmospheric radiation, and other factors, which usually lead to the radiation intensities of non-target areas such as some scenes in the infrared image over the din point target, resulting in the phenomenon of target loss or false alarm caused by the clutter in dim point target detection. Therefore, the dim point target detection in complex background is a very difficult but practical subject. Generally, a perfect point target detection method needs the higher precision in both the hardware system and the software algorithm. To obtain a predictable and stable detection performance, system designers tend to design a constant false alarm rate (CFAR) detector, which is also quite necessary for the IRST systems. Finally, the target confirmation should be carried out in the continuous tracking of the several candidate targets. The relationship between the point target detection and the target tracking is a complementary process, which may need to be carried out at the same time. On balance, the point target detection software algorithm has higher requirements of low complexity and easy implementation in real-time systems.

Dual-band photodetectors operating within infrared (IR) and ultraviolet (UV) wavelength ranges are highly applicable for tracking and surveillance of targets applications. Because of the large band-gap and conduction band offset, nitride heterostructures are good candidates for monolithically integrated IR and UV dual-band detection. However, suffering from the considerable lattice mismatch and strong polarization effect of nitride heterostructures, photodetectors with complicated structures increase the difficulties of material epitaxy and device fabrication. Here, we present a relatively simple structure design for the detection of IR and UV signals on the same sensing area simultaneously. The responses of IR and UV signals originate from the intersubband and interband transitions in the GaN/AlN superlattice sandwiched by n-doped GaN contact layers, respectively. Experimental results show that the grown sample exhibits an absorption response peaked at 1.5 μm for TM-polarized input lights. Meanwhile, the prototype sample also has a strong photocurrent response at wavelengths shorter than 350 nm, which is mainly decided by the band-gap of the GaN/AlN superlattice. These results prove the feasibility of the proposed structure of detecting IR and UV dual-band signals.

Recently, type-II superlattice (T2SL) infrared detectors have drawn a lot of attention. Compared with II-VI-based HgCdTe materials, III-V-based T2SL materials (InAs/GaSb and related alloys) have higher quality, uniformity and stability. Besides, T2SL infrared (IR) detectors have flexibility in energy-band engineering. T2SL IR detectors have higher theoretically-predicted performance than HgCdTe IR detectors, and are commonly considered to be the most promising alternative for the state-of-the-art HgCdTe IR detectors. T2SL IR detectors have experienced significant progress over past few years, in areas of material epitaxy, band structure design, and device processing methods. On basis of summarizing and analyzing literature recently published, this paper presents the development history, current status and advanced technologies of T2SL IR detectors. We firstly introduce the T2SL material, working principle and its advantages. Then we review several band structures and advances of T2SL IR detectors in some famous research institutes. Finally, advanced T2SL technologies are presented, such as HOT (high-operating-temperature), dual-color and small SWaP (Size, Weight and Power) T2SL IR detectors.

In this paper, we designed a novel tunable VCSEL structure with a top wave-like mirror for polarization control. Based on rigorous coupled-wave theory, the reflection characteristics of the top wave-like mirror structure for different polarization modes was analyzed. The wave-like structure was designed and optimized to control the output polarization mode of tunable VCSEL. The results show the threshold gain of TM mode is always greater than the TE mode during 84.5nm wavelength tuning range, the maximum increase is more than 10 times. It is indicated that the designed structure of tunable VCSEL can effectively realize stable polarization mode.

The EMVA1288 Standard is mostly used for testing and selecting area array and line array image sensors. Nowadays, Time-Delay-Integration (TDI) image sensor, especially novel type TDI-CMOS, will be widely applied in high-quality and low noise imaging domain. The purpose of this paper is to give analysis and improving the test of photo-electricity parameters of TDI image sensor. TDI image sensor has its particular architecture, so it is provided with some unique parameters, such as charge transfer efficiency (CTE). Therefore it is significant to research the improving evaluation method of EMVA1288 Standard. According to the TDI mode, modeling of TDI pixel based on linear model and noise model is achieved. Then researching and finding the improved methods to test CTE and system gain. The measured results of a device are given. The application of evaluation method in this paper is helpful in the process of developing and selecting TDI image sensor.

This paper presents a new design method of infrared search and track systems, that is, optical-digital joint optimization design method for IR detection of unresolved target. It is different from the traditional design concept of optics-detector-preprocessor sequential and independent optimization. The merit function suitable for infrared unresolved target detection is constructed for the whole infrared search and track system, and the design parameters of optical lens, infrared detector and digital preprocessor are taken as optimization variables, and then the global optimization is carried out through effective algorithm to find the overall optimization results. The proposed method belongs to an advanced form of computational optical imaging, which relaxes the local design constraints, and has a greater degree of design freedom. The research results show that it is significantly better than the traditional method.

The application scene of convolution neural network is more and more extensive, which can be migrated to infrared field. A convolutional layer accelerator is designed on the FPGA to meet the needs of miniaturization and low power consumption of embedded devices. The author reduces the model about 4 times by low-bit quantization,reduces the invalid calculations through padding processing,improves computing efficiency through data flow and parallel computing, effectively reduces the computation time of the convolution layer. Ultimately, taking the SSD algorithm as an example in the FPGA, the author reduces the calculation time to about one tenth of the cpu calculation time. At the same time, the decrease degree of the macro detection result mAP50(mean average precision) caused by quantification is within 3%, and the decrease degree of detection rate and false alarm rate is within 1%.

In the polarization imaging of shortwave infrared sea surface background, the sea surface flare as a strong interference source seriously reduces the contrast of the image, resulting in low probability of target detection, high false alarm rate, easy to be lost in tracking and other problems, which must be suppressed. The traditional solar flare suppression method suppresses the s component of the solar flare by adding a horizontal polarization filter, ignoring the p component, resulting in a general effect on flare suppression, and can not establish a relationship with the position of the sun and the position of the detector. real-time suppression of solar flare. To solve this problem, a real-time method to suppress solar flare is proposed in this paper. Combined with Cox-Munk model and polarization bi-directional reflection distribution function (pBRDF), the rough sea surface flare reflection model is established, and the polarization filter is adjusted according to the polarization angle of the flare. The related polarization degree and polarization angle images are simulated by MATLAB, and the results are in good agreement with the reality, which verifies the accuracy of the model.

Infrared imaging spectrometer can provide scene image information and spectral information at the same time, so as to deeply analyze the components and characteristics of the scene target. Due to the low resolution of the existing long-wave infrared imaging spectrometer filter and dispersion devices and the serious attenuation of signal energy, the time-modulated Fourier transform infrared spectrometer has a large volume and a high cost. In this paper, we propose a compact snapshot-type long-wave infrared computational spectral imaging method, which provides a new method for infrared spectral imaging and target recognition technology. Based on the coded aperture snapshot spectral imager (CASSI), we propose an imaging method that shares the main lens with two optical paths. One optical path is mainly composed of a coded mask, a relay lens, an amici prism, and a long-wave infrared detector. Its spatial and spectral resolution is determined by the encoded mask and the dispersive element. The optical system finally obtains an aliased two-dimensional image on the detector. The other optical path uses a long-wave infrared detector to provide high-resolution spatial information. Combining the two paths to obtain high-resolution infrared spectral image information through a compressed sensing reconstruction algorithm. The new spectroscopic imager described in this paper has the advantages of real-time detection, long-distance monitoring, and high sensitivity. It is especially suitable for mobile platforms of unmanned aerial vehicle and NanoSat. Can be widely used in trace gas detection, environmental pollution monitoring, medical diagnosis and military aircraft identification and guidance of anti-missile.

A novel laser heating method to heat specimen rapidly and uniformly to1800K in spectral directional emissivity of infrared stealth materials measurement is proposed, and a measurement system is developed based on this method. The measurement system is mainly composed of a laser heating device, a specimen heating bin, a reference blackbody, and a infrared spectrum radiation detecting device. The spectral directional emissivities of one kind of aluminum alloy specimen at room temperature and high temperature are measured using the measurement system, every measurement is completed in 3 minutes. The results indicate that spectral directional emissivity of infrared stealth materials at high temperature can be measured using the measurement system rapidly and accurately.