This PDF file contains the front matter associated with SPIE Proceedings Volume 12534, including the Title Page, Copyright information, Table of Contents, and Conference Committee information.

In my career, I have worked with several cryogenically cooled infrared detector technologies over four decades. During this period, I worked with extrinsic silicon, InGaAs, quantum well infrared photodetectors (QWIPs), quantum dot detectors, and type-II superlattice (T2SL). I am also aware of many infrared detector technologies though I haven’t directly worked on them. In last several years I was involved with managing, fabricating and delivering infrared focal planes for cameras and observational instruments. In recent years, I am involved in developing infrared cameras, multi-band, and hyperspectral instruments for Earth and planetary observational applications. In this presentation, I’ll discuss my experience as a technologist in developing novel infrared detectors/ focal plane arrays for various remote sensing applications and how T2SL focal plane array technology has changed the Infrared Landscape.

The short wave infrared (SWIR) spectral band is an emerging domain thanks to its large potential. Close to VISible/Near Infrared wavelengths, SWIR images interpretation is made easier for the users. In this spectral region, new opportunities can be found in several fields of applications such as defense and security (night vision, active imaging), space (earth observation), transport (automotive safety), or industry (nondestructive process control, food and plastic sorting). In the frame of this paper, the development of a small pitch InGaAs technology for the SWIR band is described. The objective is to fuel an increase in focal plane arrays (FPA) resolution with the development of a10μm pitch process. A specific Read Out Integrated Circuit (ROIC) has been designed to address a high frame rate and various functions like the selection of lines or specific windowing. The ROIC exhibits 3 different gains. A full set of electro-optical characterizations for a VGA 10μm pitch focal plane array is presented.

Attollo Engineering specializes in developing small pixel pitch infrared imaging sensors. Small pixel pitch sensors offer opportunities for imaging system SWaP reduction that open up a variety of SWaP-constrained applications that were not previously feasible. The size and weight of the optics also decrease because of the small pixel, enabling further SWaP savings. Attollo Engineering will present characteristics of their SWIR camera line, which includes QVGA, VGA, and HD formats, all using a 5 µm pixel pitch sensor based on InGaAs detectors. They will also show imagery taken with their SWIR cameras.

Event-based imagers are an emerging class of sensors with demonstrated advantages relative to traditional imagers. Event-based vision sensors have a limited number of output bits that are only responsive to image variations, thus overcoming the speed and power constraints of the conventional imagers based on image integration. The SWIFT EI the first event-based InGaAs sensor that is sensitive in the Visible to SWIR band (600–1700nm). The main novelty of this sensor is that the event channel outputs in parallel to a conventional fast imaging channel. Moreover, we can reconfigure the event channel to provide a fast laser pulse detection mode (3^rd generation ALPD), which also outputs in parallel to the integrated image. In that context we described in detail the architecture, key features, and preliminary simulations of the ROIC. The SWIFT EI is a low SWaP product optimized for tactical wide distribution applications that incorporates the event based FPA. We will present measurement results of the high frame rate (HFR) imaging channel; the event channel that can reach up to 25 kHz of negative, null, or positive signal, and the laser detection channel providing a single bit detection frame up to 50 kHz. This feature is ideal for multi spot tracking and pulse repetition frequency (PRF).

Shortwave infrared (SWIR) sensors are a key component of modern-day laser designation and surveillance systems; allowing the warfighter to covertly acquire and target adversaries with common infrared laser wavelengths including 1064nm and 1550nm. SWIR Vision Systems builds high-resolution sensors using colloidal quantum dot (CQD) photodiodes sensitive across the spectral band from 400 to 2000nm. CQD technology provides the ability to image common designation lasers while simultaneously offering the benefits of high-pixel-count imagers – including wide fields of view and more pixels on target. To demonstrate the utility of the CQD detector technology for this application, SWIR Vision has carried out tests imaging short pulse lasers in outdoor environments using its Acuros HD camera. These tests were carried out side-by-side with a commercially available InGaAs camera. This paper will show the results of this testing and provide a comparison of the performance, range, and pulse-power requirements for designation applications using CQD and InGaAs technology. As the importance of SWIR laser designator systems in the modern battlespace continues to grow, SWIR CQD technology shows promise to differentiate and maintain an advantage over our adversaries both with higher resolution imagery as well as the ability to detect wavelengths longer than what has traditionally been possible by traditional IR detector technologies.

III-V and II-VI compound semiconductors have long dominated high sensitivity IR detection applications. In recent years, technologies such as strained layer superlattices have emerged, but they still require cooling. We have recently demonstrated uncooled detection in predominantly the SWIR with residual detection in the MWIR and LWIR. The devices operate at uncooled, ambient temperatures and exhibit mA/W responsivities when measured against a cavity blackbody radiator, have low-kHz -3 dB electronic bandwidths, but with specific detectivities comparable to III-V and II-VI semiconductors. In one example, a polymer with an undoped SWIR bandgap well beyond 1 μm—a first for any polymer system—demonstrated broadband responsivity of 6.5 mA/W against a cavity blackbody radiator. An introduction to conjugated polymers is included that provides a bridge between the polymer and semiconductor communities.

Recent short-wave infrared (SWIR) sensors have demonstrated in-pixel multimode capabilities. One of the additional modes is range finding. High resolution range finding is increasingly becoming vital functionality in high precision targeting and imaging systems. Highly precise and accurate range-to-target information is essential for many modern commercial and military applications. With the recent advances in LiDAR (Light Detection and Ranging) technology, range measurement accuracies as low as a centimeter at kilometer ranges. Sensors Unlimited Inc. (SUI), a Raytheon Technologies (RTX) Company, has been developing these multimode sensors using traditional PIN-based InGaAs detector technology. However, the capability of these sensors has been extended through the introduction of Avalanche Photodetector (APD) InGaAs sensors. This APD technology has been developed onshore to better serve the onshore community requiring simultaneous laser tracking, ranging, and imaging applications. In this work, SUI offers an update on previously presented, PDA-specific development, most specifically related to the advancement of Geiger Mode Avalanche Photodetectors (GMAPD). SUI’s APD technology is in direct response to the challenging SWaP and NEI performance requirements of active imaging and tracking applications. This update includes 2D and 3D TCAD simulation results with a comparison with measured performance results. Finally, initial results related to the advanced development of Geiger Mode Avalanche Photodetectors (GMAPD) themselves as well as supporting electronics is given. The revitalization of SUI’s APD development is a direct response to the challenging SWaP and longer-range with higher accuracy performance requirements of active imaging applications. SUI’s most recent APD design improvements facilitates greater signal to noise ratio at the pixel, which subsequently enables a supporting ROIC pixel design with improved performance.

Improved situational awareness using SWIR is of upmost importance to many of SUI's, a Collins Aerospace company, customers. SUI’s proprietary Asynchronous Laser Pulse Detection (aka MMT, Multi-Mode Tracking) methodology facilitates the detection, tracking, and decode of NIR-SWIR lasers at the pixel level at all ambient lighting conditions without the compromising imaging performance. This paper describes advancing this technology for improved hostile detection. The main challenge of hostile detection using infrared imaging technologies is the requirement for fast frame rates as the signal lifetime is short for both retroreflection and muzzle flash detection. In this work, SUI demonstrates how its MMT technology can be leveraged for low SWaP hostile detection applications.

Sensing and imaging applications such as light detection and ranging (LiDAR) require a receiver with highly sensitive photodetectors such as avalanche photodiodes (APDs), which have high gain and speed. Conventionally, APDs use mesa structures that limit device scaling due to increased surface leakage current at smaller sizes. Planar-architectured APDs address high surface leakage current by reducing the exposed surface of the devices. They also eliminate complicated mesa fabrication steps such as mesa etching and passivation. Fabrication of APDs with planar architecture conventionally uses processes such as ion-implantation, which can damage the epitaxially-grown material, and metal organic chemical vapor deposition (MOCVD), which can be expensive and time-consuming. In this paper, we assess an alternative fabrication process to make APDs using Zn diffusion into epitaxially-grown material stacks through atomic layer deposition (ALD) of ZnO. In particular, we have explored Zn diffusion in epitaxially grown type-II InGaAs/GaAsSb superlattice on InP substrates. ZnO was grown with various thicknesses on the epitaxial stack using ALD, and the Zn was diffused into the stack using a diffusion furnace under forming gas. The effects of the diffusion process were investigated using various characterization techniques, including x-ray diffraction, and secondary ion mass spectroscopy. The optimized Zn-diffusion process is then used to make planar diodes. Further optimization of this diffusion process may lead to a rapid, manufacturable, and cost-effective method of developing planar APDs.

Trends at detector and system levels are mainly linked to reduced SWaP (Size, Weight and Power), higher availability and higher flexibility of integration. As a first example, originally, the detectors used to be cooled around 77K. More recently, depending on the detector technology, the required bandwidth, and the performance, detectors operating temperature may vary in a broader range between 60K to 170K. In order to avoid having to develop a cooler for each specific detector, it is therefore necessary to have coolers that are capable to operate on a wide range of conditions. This flexibility shall also improve the SWaP at system level by reducing the electronic driver size and by making easier the cooler integration in the system. Moreover, the cooler reliability is a major performance requested by customers, especially for 24h/7 applications, which are a growing market. Indeed, most of the time, coolers are one of the components dimensioning the lifetime of the system. Several improvements were made on rotary coolers in order to fill the reliability gap between linear and rotary coolers. It is important to continuously improve this key parameter. All of these trends were taken into account in Thales rotary coolers’ road map in order to propose on the market coolers able to answer to these new users needs. The paper will focus on the last works made at Thales on rotary coolers in the frame of that target and more particularly on the RM2 cooler. Firstly, we will illustrate the latest reliability improvements made on this cooler with lifetime tests results. Then, we will describe a new option for this cooler allowing us to have better SWaP and flexibility.

Technological innovations advance the state of the art for low size, weight, and power (SWAP) cryocoolers for high operating temperature infrared imaging. The authors present essentials of mechanical design, outcomes of theoretical modeling/optimization and initial testing for a cost-effective low SWAP split Stirling crycooler.

In 2007 AIM introduced SF100 Moving-Magnet Flexure-Bearing compressor suitable for Stirling and Pulse Tube coldfinger. In this paper there will be an update on several improvements for SF100 Stirling-Type Cooler suitable for SADA-applications. The improvements to both compressor and regenerator focus on improvements regarding: - Compliance with harsh environments - Reliability - Efficiency - Increase in power density, i.e. max. input power - Fast cooldown, especially at high ambient temperatures Design approach and performance data as well as qualification results will be shown.

Over the past few years, the available literature shows signaled great progress in improving the performance of ultraviolet and infrared detectors, even exceeding the physical limit of detection determined by signal and background fluctuation limits. This is particularly true for Ga₂O₃ phototransistors and a new generation of photodetectors with active areas containing low-dimensional solids. The purpose of this paper is to show cases of overestimation of the performance of ultraviolet and infrared detectors noted by the author and to draw attention to their fundamental physical limitations which are overlooked in the characterization of detectors.

The increased performance and versatility of epitaxially grown III-V Type II strained layer superlattice (T2SLS) materials led to disruption of the IR sensor industry over the last decade. Increased performance at reduced cost are highly attractive to many end customers. Access to this detector material combined with the L3Harris front-side illuminated FPA structure lend to mechanically robust FPAs with array size only limited by source detector wafer size. This paper will present an overview of the technology development at L3Harris and the progress that has been made introducing SLS material into high operating temperature (HOT) MWIR sensors, LWIR sensors and dual-band (MW/MW & MW/LW) sensors.

In this paper, the main parameters contributing to the high-imaging quality of IRnova’s Type-II superlattice (T2SL) detectors are studied: the sensitivity, uniformity, modulation transfer function (MTF) and random telegraph signal (RTS). High-sensitivity and good uniformity across the detectors are displayed by their low-noise equivalent temperature difference (NETD) with average values of 22 mK (@ F/5.5, 125 K) and very narrow histograms. MTF values very close to the ideal ones have been demonstrated for Oden MW detectors, reaching 0.6 at the Nyquist frequency, along with a single-digit number of flickering pixels. Furthermore, thanks to the optimization of design and fabrication processes of Oden MW detectors, diffusion-limited dark current densities have been reached that are independent of the perimeter/area ratio for the pitch sizes down to 10 μm. Leveraging this achievement is the upcoming Njord MW HD detector (1280×1024 pixels, 10 μm pitch), for which the bias- and temperature-dependence have been assessed using IRnova’s in-house simulations tools. NETD values lower than 25 mK are shown for temperatures up to 140 K. The simulations were based on measured bias- and temperature- dependencies of the detectors on 10 μm pitch, combined with parameters of the dedicated 1280×1024 ROIC on 10 μm pitch design.

ASELSAN, the largest defense company in Turkey, develops high performance electro-optical systems for various applications. Due to the technical challenges and relatively higher cost associated with MCT technology, there is a search for an alternative material system all over the world, especially for the applications that need relatively low quantum efficiency and high volume production. Research and development activities have been carried out on developing InAs/GaSb T2SL detectors for mid-wavelength infrared (MWIR) technology in ASELSAN. In this paper, recent results for 15μm pitch MWIR T2SL and 10μm pitch MWIR T2SL detectors are presented. InAs/GaSb T2SL epilayers are used for FPA fabrication with mesa pixel structures. Typically, over 99% operability and less than 25 mK NETD values are achieved for 15 μm pitch 640x512 format MWIR T2SL FPAs at IDDCA level (F/4). Besides ongoing efforts on development of FPAs with 15 μm pixel pitch, development activities for pixel pitch reduction are also initiated recently for MWIR T2SL FPAs for improving resolution and promising results are achieved. For this purpose, MWIR T2SL FPAs with 1280x1024/10μm resolution/pixel pitch are fabricated and NETD values less than 25 mK (F/2) with pixel operability of 97% are achieved for the first prototypes.

L3Harris has developed a multispectral sensor architecture that opens doors to the incorporation of enabling functionality for future applications. This sensor architecture, while detector material agnostic in nature, builds upon our type-II superlattice (T2SL) technology, which was originally developed through the tri-service Vital Infrared Sensor Technology Acceleration (VISTA) program. The sensor architecture consists of stacked absorber layers that can be individually processed as they are integrated into the sensor stack. This layer-based approach enables the insertion of additional functional structures to enhance performance, such as embedded dielectric filters for efficient spectral separation, which would not be possible in a monolithic design. Due to the additional degrees of freedom with this configuration, it is necessary to design the filter such that it is compatible with the process required to functionalize each absorber while simultaneously meeting performance requirements. In this work, we report on various filter designs impacting the performance of focal plane arrays (FPAs) having embedded dielectric filters, under various practical constraints.

Ga-free InAs/InAsSb T2SL XBn detector is now a reliable candidate for high-performance focal plane arrays in the MWIR (3-5μm) domain. However, this T2SL is a very anisotropic quantum structure having a type-IIb band offset alignment where electrons are rather delocalized all over the structure while holes are strongly confined in deep InAsSb quantum well. This configuration could penalize the absorption and the hole minority carrier transport but MWIR detector device without anti reflection coating shows quantum efficiency higher than 50%. Considering results of specific measurements and band structure calculation, possible carrier transport scenario is presented to explain such performance.

Each MWIR and LWIR detectror, which is widely used in various civil and military including chemical identification, atmospheric monitoring, guided weapon and surveillance reconnaissance systems, is advantageous for detecting hot and cool targets, respectively. Dual-band or multi-band detector that is able to detect more than two bands with only single detector has excellent recognition and identification capablities. Therefore, various groups have studied dual-band or multi-band detector since 1998. In this work, a 20 μm 640×512 dual-band midwave and longwave infrared detector with nBn structure was studied. A nBn detector is not only effective in reducing dark current, but it is relatively simple to implement a dual-band detector by growing MWIR and LWIR absorber layers on both sides of a barrier layer. The dual-band detector acquires each MWIR and LWIR bands by selecting the applied bias direction. Consequently, the 20 μm 640×512 dual-band MWIR/LWIR FPA hybridized to read-out integrated circuit (ROIC) exhibited that an average noise equivalent temperature difference (NETD) and operability of both MWIR and LWIR modes were less than 25 mK and more than 99.5 %, respectively.

L3Harris Technologies recently patented infrared detector arrays made by directly bonding a III-V Type II Superlattice (T2SL) detector to an optical silicon wafer. The process is performed at the wafer level and can replace L3Harris’ traditional adhesive bond. Direct bonding offers several advantages over adhesive bonding. First, higher operability and fewer defect clusters result due to the rigorous preparation of wafer surfaces before bonding. The number of antireflection coated surfaces is decreased from three to one. Higher quantum efficiency results from the improved transmission through the III-V/Si interface. This is critical for multi-band detectors, which operate over a relatively wide spectral bandwidth. Finally, the bond withstands higher processing temperatures than the adhesive bond. The process has been demonstrated on 150mm silicon with III-V wafers up to 125mm in diameter. The optical silicon remains in the finished focal plane array (FPA), serving as a protective window on the front surface of the delicate detector array. The silicon is a key component to L3Harris FPAs, forcing a coefficient of thermal expansion (CTE) match between the detector array and the silicon readout integrated circuit. The CTE match enables large array formats and long thermal cycle life without shimming.

Unmanned airborne and dismounted soldier capability requirements continue to push for reduced size, weight, and power (SWaP) and high sensitivity infrared (IR) imaging in applications that were not previously practical. In response to these needs, Attollo Engineering has developed a 1280x1024, 5μm pixel pitch cooled mid wavelength infrared (MWIR) sensor that pushes the envelope in pixel pitch in addition to a 1280x1024, 10μm pixel sensor dual band sensor with additional sensitivity in the short wavelength infrared (SWIR) in order to exploit SWIR phenomenology including laser see spot functionality. Both of these sensors offer MWIR sensing capabilities but are also able to leverage aspects of Attollo’s detector design to enable SWIR sensing to varying degrees. This class of small pixel cooled, single and dual band IR sensor technology represents advancements in all aspects of the sensor’s design and development, and we will discuss the innovations made at Attollo to enable this capability including epitaxial detector design based on III V compound semiconductors, detector array and focal plane array fabrication, design of a low noise, dual band CTIA/DI readout integrated circuit (ROIC), vacuum dewar packaging, and electronics and firmware design. In this paper we will present on the status of high definition small pixel pitch MWIR and dual band SWIR/MWIR imaging technology at Attollo as it relates to these sensors including design and measurement data and imaging.

Building on our previous efforts, IQE will demonstrate the high-volume manufacturability of our MBE growth process for GaSb-based MW and LW infrared photodetectors (IRPDs) structures. This work focuses on the ability to scale across multiple MBE reactor platforms and across multiple production sites. Supporting the scaling of FPAs requires the ability to grow IRPD structures on larger diameter substrates with good wafer-to-wafer and cross-platen uniformity. Enhancement of Ga-free LWIR QE may be realized through metamorphic growths and the use of high-index substrates. We have also extended our heteroepitaxial growth effort on Si substrate from bulk InAsSb to Sb T2SL MWIR structures.

HgCdTe (MCT) is a material system intensively used for IR sensing from space as it offers high detection performances. It is also considered as a relatively radhard material system as the performance degradation due to radiative space environment are usually not dramatic compared to other material systems. However, there is no clear understanding of the effect of displacement damage dose (DDD) or even total ionizing dose (TID) in this material system. In this communication, we present an extensive study of the damage due to protons on n/p MCT diodes sensing in the MW range. Both FPA full arrays, single diodes and MIS test structures have been irradiated with 63 MeV protons up to relatively high doses (8e11 protons/cm²). We will report about the evolution of dark current and noise degradation (mainly due to RTS), during and after irradiation, as well as after a thermal cycling up to 300K and even after a recovery annealing at higher temperature.

One limitation of dark current for HOT HgCdTe MWIR detector is diffusion current related to Auger generation mechanisms. However, this limit can be overcame using a fully depleted P-i-N diode structure. On the way toward PIN structures, we discuss on the impact of Auger suppression on partially fully-depleted photodiodes. Indeed, we report dark current evolution with temperature following a diffusion trend line but also a dark current decreasing when increasing the applied voltage in the temperature range 120-180K on the same detector. If the SRH current in the space charge region is very low, an Auger suppression mechanism should result in a decrease of the total dark current when depleting a larger volume of the pixel on such detector. This hypothesis could match with the reported dark current data obtained on detectors manufactured at CEA LETI and Lynred. They are based on p/n diodes, 7.5μm pitch arrays connected as 2x2 super-pixels to a 15μm pitch ROIC (640x512 VGA format).

Pixel pitch reduction as well as high operating temperatures (HOT) of IR focal plane arrays (FPA) have attracted huge development effort in recent years. While the development journey is continuously moving towards an ultra-small pixel pitch of ~5μm for MWIR FPAs by several manufacturers, for their LWIR pendants the focus has not been on pitch size reduction and HOT so far. This comes as a surprise, since low SWaP (size, weight, and power) LWIR modules in combination with high spatial resolution are of great interest due to their superior electro-optical (e/o) performance in particular for scenarios with degraded visual environment. An SXGA-format (1280x1024 pixels) MWIR module using a 7.5μm pixel pitch is currently under development at AIM. Excellent performance data were obtained with the first prototypes up to elevated operating temperatures of ~150K thanks to the AIM HOT MCT p/n detector technology. SWaP characteristics are analyzed for both a compact configuration using the in-house ‘SX035’ cooler and for an even smaller variant referred to as ‘HOT Cube SX’. Moreover, a low SWaP VGA-format MWIR module is under development at AIM using a 10μm pixel and a HOT rotary cooler. First measurement results will be shown. Promising first e/o data of XGA-format 10μm pitch LWIR FPAs manufactured in a planar MCT p/n technology have already been presented by AIM last year. Recently, the performance has been improved which is underlined by a high thermal resolution of ~30mk and a low level of noisy pixel at HOT of ~100K. Latest results as well as key features of the possible low SWaP IR-module configurations will be discussed.

The quality of surface finishing of CdZnTe wafers plays an essential role for the development of high-performance HgCdTe detectors. The surface processing of CdZnTe wafers is regarded as one of the most challenging tasks due to the soft-brittle material characteristics of CdZnTe crystals. The HgCdTe-based detector technology requires CdZnTe wafers with atomatically smooth surfaces because of the HgCdTe epitaxial growth. The wafer processing cycle consists of multiple steps starting with as-cut wafers from in-house grown CdZnTe boules followed by a series of consecutive polishing processes. CdZnTe wafers are then cleaned prior to wafer characterization and final inspection. The objective of this work is to demonstrate the current status of CdZnTe wafer processing technology at Aselsan through the as-cut wafers to the polished wafers promoted as finished products. The current wafer processing technology produces in-house grown epi-ready CdZnTe substrates with low surface roughness (<0.5 nm) and low flatness (<1 μm) in a repeatable fashion.

ASELSAN, the largest defense company in Türkiye, develops high performance electro-optical systems for various applications. Research and development activities have been carried out on developing HgCdTe (MCT) detectors for long-wavelength infrared (LWIR) and mid-wavelength infrared (MWIR). In this paper, recent results for VGA 15μm pitch MWIR MCT detectors at IDDCA level are presented. P-on-n MCT epilayers are used for FPA fabrication with either mesa or planar pixel structures. Typically, over 99% operability and less than 20mK NETD values are achieved for 15μm pitch 640x512 format MWIR MCT FPAs at IDDCA level (F/4) in a repeatable fashion. Thermal cycle, mechanical shock, vibration and environmental tests (such as storage and operation under hot and cold temperatures) were applied to these MWIR MCT IDDCAs and passed successfully. Besides ongoing efforts on development of FPAs with 15μm pixel pitch, development activities for pixel pitch reduction also initiated recently for MWIR MCT and very promising results are achieved.

Infrared Imaging has played a vital role for US Army for tactical ground and air operations over the previous decades. High-confidence image-based scene segmentation and object detection will be foundational in increasing the tempo of operational decision making for Army scouting, surveillance, and threat detection systems. New hardware and software solutions to increase the first-look confidence of imaging sensor systems, and the self-awareness of these sensors to their own limitations, will be critical for the success of future Army air and ground system programs. This paper will review current and future technical challenges and priorities for Army tactical imaging sensors for operational autonomy.

Cyan Systems has continued to mature our small pixel camera performance, including improvements in the packaging, optics, and electronics. The associated camera components demonstrate key resolution and enabling capabilities. We report on recent results from our new digital readout integrated circuit (DROIC) small pixel mid-wave infrared camera with an ultra-high-definition (3840 x 2160) format, in addition to demonstrations with Cyan’s CS-3 full-high-definition (1920 x 1080) camera. We address small pixel spatial sampling and modulation transfer function issues as the pixel size shrinks, and we examine the difference between the performance of present devices and the new generation of small pixel cameras.

The latest generation of high-definition small pixel pitch sensors require a new generation of optics to realize the full performance of these detectors. We will describe a modular architecture for Mid-Wave InfraRed (MWIR) continuous zoom cameras and some novel features that our approach enables.

This work reports on the results of shore-based maritime environment and shipboard at-sea testing of the Shipboard Panoramic Electro-Optical Infrared (EO/IR) Cueing and Surveillance System (SPECSS) developed under an Office of Naval Research (ONR Code 31) Future Naval Capabilities program. The key technology enablers for the system are 25 Megapixel mid-wave infrared (MWIR) cameras that utilize III-V strained layer superlattice nBn detectors, and highly-parallel detection processing on the latest commercially available graphic processing units (GPUs). The effects of the maritime atmospheric and background conditions, as well as ship motion and line of sight stabilization, on the performance of the system will be discussed. Demonstration of highly accurate angular target cueing to other high resolution sensors will be presented.

In this paper we review the optical design challenges and performance of a new NIR+SWIR zoom lens. The main features of this lens include a wide spectral range of 700nm to 1700nm covering both NIR and SWIR bands, a 10x zoom ratio; a 25-250mm EFL range, a F/5.5 to support both 10um and 5um pixel pitch detectors including SXGA format, a low SWaP athermal design, and a high MTF performance along the zoom range and over a wide temperature range. The most significant challenge in this design was the wide spectral range requirement of 700nm to 1700nm, resulting in challenging control of the chromatic aberrations. Another challenge was the reduced SWaP requirements which limited the physical length of the design and therefore increased the performance sensitivity to tolerances. These design considerations and the high-performance results over the full zoom range are described.

System requirements for IR imagery continues to push the limits to smaller pitch and larger formats. This enables better resolution at distance and more situational awareness. L3Harris continues to reduce pitch and increase the array size for Type II Strained Layer Superlattice (T2SLS) FPAs. Access to this detector material combined with the L3Harris front-side illuminated FPA structure lend to mechanically robust FPAs with array size only limited by detector wafer size. The current epitaxially grown material is highly uniform across the entirety of the 125 mm wafer. L3Harris will present results from improvements demonstrated on the 6144 x 4096 focal plane with an 8 micron pitch. These improvements include higher operability, yield and lower noise.

LYNRED is oriented towards excellence in II-VI, III-V and bolometers technologies, covering all Society’s needs in term of infrared detection. Our vision is to preserve and protect, and more than ever, our goal is to provide the right technology to the field missions, spatial and industrial applications, and more generally the right technology to customers’ needs. For this purpose we are developing for the next generation pitch, MCT HOT technology, for extended MW band as well as III-V HOT MW blue band technology. Many challenges have to be addressed for future small pitch, large format and HOT detectors. Electrical and optical crosstalks as well as image quality and stability, are one of the prime concern for detectors with pixel pitch below 10μm. We will discuss about the trade-off between the different material properties and detector performances to ensure mandatory minimization of Minimum Resolvable Temperature Difference (MRTD) for range optimization. We will then review latest 7.5μm pitch development at LYNRED, with SXGA formats, based on II-VI and III-V HOT materials, in terms of operability, residual fixed pattern noise (RFPN) and Modulation Transfer Function (MTF) optimizations.

Today, the trend line in cooled infrared detector manufacturing is to go towards very small pixel pitches and to operate at high temperatures. To take full advantage of this pitch reduction, the modulation transfer function (MTF) has to be correctly measured to optimize the development of HgCdTe p/n planar diodes, where the MTF is degraded due to the large diffusion length of minority carriers. Furthermore, as the pitch decreases close to the wavelength, the measurement of its MTF becomes difficult using traditional optical projection methods, which are intrinsically limited by diffraction. In order to assess this MTF, an original characterization method is therefore investigated at LETI, involving an electron beam instead of an optical beam to excite the pixels: the electron beam induced current (EBIC). This method introduces a low MTF degradation (estimated by Monte Carlo simulations) on the measurements that can be neglected, thus giving access to a direct estimation of the pixel MTF, with no need for deconvolution. This work shows EBIC measurements carried out on HgCdTe planar diodes with small pixel pitches: 7.5μm and 5μm. The resulting MTF is compared with MTF computed by finite-element modeling using an electronic and photonic excitation.

MCT p-on-n photodiodes manufactured at Lynred and CEA-LETI have demonstrated state of the art performances for HOT applications. On blue and red mid infrared bands on 15μm pixel pitch, respectively 150 and 130K operating temperatures have been obtained, due to diffusion limited dark current and low defectivity. To achieve equivalent results on smaller pixel, the p-on-n technology at DEFIR, joint laboratory between Lynred and CEA-LETI, has been improved. The technological process was modified to ensure a proper diode formation and to efficiently passivate the interface between MCT and encapsulation layers, especially in the vicinity of the space charge region. The manufactured arrays with a 5.3μm cutoff wavelength have been hybridized on a digital output SXGA (1280×1024) direct injection ROIC with a pixel pitch of 7.5 μm. This paper present the measured current, blackbody responsivity and RMS noise on FPAs with F/4 numerical aperture. We will also discuss spectral response, quantum efficiency, shot-noise limited photodiodes and noise histograms shapes and their distribution tails at 130K. The very low number of defective pixel allow to address higher operating temperature and measurements have been performed at 140K and even 150K with very limited performance degradation. Pixel pitch of 5 μm has been characterized on test chips and present I-V curves with low dispersion and long bias plateau. As for larger pixel sizes, these photodiodes are shot-noise limited. Modulation transfer function has been measured by electron beam induced current and presents high value, up to 56%.

Modern airborne platforms are generally equipped with Missile Approach Warning System (MAWS) based on high end Infra-Red (IR) detectors. The airborne platforms put high demands on the IR detectors, both for electro-optical performance and for durability against harsh environmental conditions. Recently, these systems have evolved to also include situational awareness for the pilot, and hence are defined as Electro Optical Distributed Aperture Systems (EO-DAS). Detectors for both MAWS & EO-DAS require a large Field of View (FOV) in order to cover a solid angle of 4π steradians around the platform. However, while MAWS can be satisfied with standard resolution sensors, a high resolution detector is a must for situational awareness,. This requirement is met by large format detectors (>1 Mega pixel) having a smaller pixel pitch. In addition, a high frame rate is required for preventing image blur and for reducing the False Alarm Rate (FAR). Thus in order to maintain sufficient sensitivity and a low FAR, the IR detector must be sensitive to the full Mid-Wave IR (MWIR) band of radiation. Since MWIR detectors are cooled to cryogenic temperatures, the combination of a large format and a high frame rate is quite demanding. Furthermore, the conditions in most aerial platforms are harsh, with very high environmental temperatures, a high shock level, and significant levels of vibration. Therefore, the detector Dewar must be ruggedized and the cooler must maintain a large enough cooling capacity, even though the volume available for the detector is usually very limited. To address all of the above requirements is becoming extremely challenging. Over the years, SCD has developed several MAWS solutions that have been successfully integrated into systems and combat proven. In this paper, we review these solutions, and describe our recent development of a High Operating Temperature (HOT), 5 Mega-pixel detector optimized for both MAWS and EODAS requirements.

In recent years, the market for infrared (IR) sensors has expanded from traditional defense and security applications to include consumer products. As a result, there is increasing demand for embedded IR systems that integrate low-cost sensors with embedded processors. Meanwhile, deep learning has made significant advances and achieved superhuman performance in some domains. To support deep learning in embedded systems, new processors have emerged that are specifically designed for running deep neural networks. In this paper, we propose a performance evaluation method for applying quantized deep learning to low-cost IR sensors using layer-wise relevance propagation. Our method provides visualized analysis of what the neural networks learn. We demonstrate the effectiveness of our approach through experiments on a low-cost IR sensor dataset, showing that our method achieves an explainable performance evaluation method for degraded cases arising from the tradeoff between speed and accuracy of quantized detectors which is a typical problem in embedded systems with limited computational resources.

Cameras digitize real-world scenes as pixel intensity values with a limited value range given by the available bits per pixel (bpp). High Dynamic Range (HDR) cameras capture those luminance values in higher resolution through an increase in the number of bpp. Most displays, however, are limited to 8 bpp. Naïve HDR compression methods lead to a loss of the rich information contained in those HDR images. In this paper, tone mapping algorithms for thermal infrared images with 16 bpp are investigated that can preserve this information. An optimized multi-scale Retinex algorithm sets the baseline. This algorithm is then approximated with a deep learning approach based on the popular U-Net architecture. The remaining noise in the images after tone mapping is reduced implicitly by utilizing a self-supervised deep learning approach that can be jointly trained with the tone mapping approach in a multi-task learning scheme. Further discussions are provided on denoising and deflickering for thermal infrared video enhancement in the context of tone mapping. Extensive experiments on the public FLIR ADAS Dataset prove the effectiveness of our proposed method in comparison with the state-of-the-art.

This work is a review and extension of our ongoing research in human recognition analysis using multimodality motion sensor data. We review our work on hand crafted feature engineering for motion capture skeleton (MoCap) data, from the Air Force Research Lab for human gender followed by depth scan based skeleton extraction using LIDAR data from the Army Night Vision Lab for person identification. We then build on these works to demonstrate a transfer learning sensor fusion approach for using the larger MoCap and smaller LIDAR data for gender classification.

The reduction of the pixel pitch of uncooled thermal imagers is still ongoing and by the end of 2021, reached a size of only 8µm within commercially available products. In case of those small pixel sizes, design and manufacturing requirements become more challenging by facing the main technological objectives of maximizing the temperature rise within the microbolometer membrane and simultaneously, being sensitive towards the latter. Tackling these challenges, Fraunhofer IMS provides a manufacturing process for uncooled thermal imagers optimized for the LWIR regime based on a scalable microbolometer technology incorporating vertical nanotubes. Based on this technological concept, we already demonstrated microbolometers with full electro-optical functionality for pixel sizes ranging from 17µm down to 6µm. A comprehensive study of our measurements regarding the scalable microbolometer technology will be presented here. This includes a discussion about design requirements in correlation to the achieved electro-optical results covering electrical noise, NETD, thermal time constant and spectral absorption characteristics. Those key parameters will be summarized and evaluated with respect to the reduction of the active microbolometer area in case of a shrinked pixel size.

The development of a new thermometer material at Lynred has opened new opportunities towards a smaller pixel pitch. This pixel pitch reduction leads to several challenges and technological breakthroughs that are necessary both for the bolometer pixel design and the readout circuit architecture. The symposium presentation and the associated article will present a recent development at CEA-leti and Lynred of a 8.5μm bolometer pixel pitch Focal Plane Array (FPA). The drastic reduction of the pixel area leads to important changes of the micro bolometer features, especially a strong decrease of thermal capacitance Cth and increase of the thermal resistance Rth in order to maintain a high detector responsivity. The rise of the noise due to temperature fluctuation or 1/f contributions have also to be considered. The consequences of those modifications and resulting trade-offs, such as pixel fill factor and optical absorption yields in the 8-14 μm band ; pixel self-heating and Read-Out IC (ROIC) dynamic range, will be discussed. Finally, the electro-optical performances of our 640x480 FPAs are presented, and perspectives will be given.

Uncooled infrared sensor markets have grown dramatically over the past decade in various fields, including military and commercial applications. The industry has been driven by the need for high resolution in military settings and low cost in commercial markets. Pixel pitch reduction allows manufacturing of larger format and low cost production without system integration loss. Overcoming a sensitivity degradation caused by pixel size reduction is the key driver for the next generation of uncooled sensors. This paper presents a brief overview of the status of TiOx-based SXGA(1280x1024) 8μm pitch uncooled infrared sensor status with 0.18μm 3D MEMS technology and 4 channel simultaneous integration circuit to improve performance.

Multispectral uncooled infrared (IR) sensors are attracting interest for use in applications such as gas analysis, hazardous material recognition, and biological analysis. However, conventional sensors require additional filters or a resonant structure in the vertical direction. To address this challenge, we investigated plasmonic metamaterial absorbers (PMAs) such as plasmonic crystals (PCs), metal–insulator–metal (MIM), and mushroom-type PMAs. Such PMAs exhibit wavelength-selective absorption by surface plasmon resonance (SPR, and the absorption wavelength can be controlled by modifying the features of the surface pattern (e.g., the periodicity of dimples and the micropatch size). Previously, we fabricated thermopiles with PC-type PMAs and used them for wavelength-selective detection. Although PC-type PMAs have advantages such as easy fabrication, robustness against structural fluctuations, and a wide operating wavelength range, they suffer from an incidence-angle dependence and require a large absorber volume. Here, we fabricated thermopiles with encapsulated MIM (EMIM)-type PMAs using a complementary metal oxide semiconductor process. The devices consist of a top encapsulated layer, periodic top micropatches, a flat insulator layer, and a bottom reflector. The encapsulated layer functions as a protective layer. Like MIM-type PMAs, the EMIM-type PMAs exhibit incidence-angle independence and a thin absorber volume, where the top micropatch induces localized SPR with the bottom reflector, and its size mainly determines the absorption wavelength. Spectral measurements showed that wavelength-selective IR detection can be performed in the mid-IR wavelength range without any additional filters. We expect the obtained results to contribute to the realization of advanced functional uncooled IR sensors and expand their range of applications.

We report on a new manufacturing paradigm to build scalable microbolometers at the lowest cost possible. A VGA microbolometer focal plane array has been designed and fabricated using flat panel display production tools, using the Large Area MEMS Platform (LAMP). Complex MEMs sensors have been integrated with Low Temperature Polysilicon (LTPS) transistors on a Gen 3.5 glass substrate in a display production fab to realize an order of magnitude gain in the number of devices per substrate. In this modular approach, a custom readout integrated circuit (ROIC) is bonded to the glass focal plane with the Chip on Flex (COF) technique, a high volume and low-cost staple of the display industry. We discuss the potential implications of the pricing disruption to important commercial applications.

Infrared imaging arrays are well established in a wide range of industries and consumer applications. Thermopile arrays are the state-of-the-art solution for cost-effective thermal imaging at relatively low resolutions below 10 000 px. Typical applications include hand held thermal imaging, security cameras, remote temperature measurement and industrial process control. Unlike alternative approaches, thermopile arrays enable truly shutterless radiometric IR cameras. They are widely available in resolutions ranging from 8×8 to 80×64 px with all necessary signal conditioning and readout electronics (including the SPI interface) monolithically integrated on the sensor chip and with optics included. The physical size of the pixels in the array is a limiting factor for their resolution. The standard 90 μm pixels typically achieve resolutions below the mid resolution range (below 10,000 px). Here, we introduce Heimann Sensor GmbH’s new thermopile arrays† with 60 μm and 45 μm μm pixel sizes enabling an increase in resolution from the low resolution range to the mid resolution range of >10,000 px. This allows the application range to be further extended into higher density thermal imaging and lower cost security and surveillance applications, as well as to infrared computer vision.

The requirement for cooling below room temperature to limit thermal generation processes is the major drawback of infrared (IR) photon detectors. In order to obtain background limited performance (BLIP), the devices for the 3–5 μm spectral region are typically operating at T <200 K, while those for the 8–12 μm at 77 K. The current effort to produce near room temperature photon detectors (higher operating temperature conditions, HOT) are being concentrated on the interband cascade photodetectors (ICIP). That effort is mostly seen for longwave infrared (LWIR) devices operating at HOT conditions where the optimum absorber may be thicker compared to the carriers’ diffusion length, limiting the photogenerated carriers collection what leads to the decreasing of the quantum efficiency (QE) and the ultimate performance cannot be achieved in practice. The ICIP architecture allows to circumvent that issue. In addition to avoid the limitation imposed by the reduced diffusion length, the intersubband (IS) quantum cascade photodetectors (QCPs) were introduced in the early 2000s based on quantum-well infrared photodetectors (QWIPs) and quantum cascade lasers (QCLs). Those cascade detectors proved to be operating at low temperatures ~ 100 K. In this paper we present the current status of the ICIPs based on type-II superlattices (T2SLs) InAs/InAsSb n- and p-type doped active layers intended to be operating within LWIR range and HOT conditions - 210 K reached by thermoelectric coolers and room temperature. Immersed detector reached detectivity, D*~3×10⁸ cmHz^1/2/W at 300 K and wavelength, λ=10 μm.

We report on a multi-year development effort demonstrating resonant cavity enhanced photodiodes (RCE-PDs) with an all-epitaxial architecture, exploiting the GaSb matched material system and its increasingly popular family of IR absorbers. RCE-PDs redefine the relationship between signal and noise generation within IR detectors, breaking some of the performance limits constraining conventional IR detectors. High performance RCE-PDs are achieved with resonance wavelengths from the SWIR to the LWIR, using GaSb/AlAsSb DBR mirrors and InGaAsSb, InAs, InAsSb and T2SL absorbers. Finally, we introduce a graded thickness RCE-PD which can perform the function of a spectrometer on a chip.

Quantum well infrared photodetector (QWIP) technology is an excellent solution for tactical applications due to the unmatched image uniformity, stability, and pixel operability (typically <99.9%). IRnova have a high-volume production of QWIP detectors in both QVGA and VGA formats since 1999. In 2017 a VGA format QWIP with 15 μm pitch was released, which has demonstrated excellent sensitivity (expressed as noise equivalent temperature difference (NETD)) of 30 mK at F/2 and 100 Hz frame rate. QWIPs also enable compact designs for polarimetric imaging by implementing polarization-sensitive lamellar gratings directly onto the detector pixels. Such polarimetric LWIR detectors (peak absorption at 8.5 μm) have been fabricated at IRnova as both QVGA and VGA format FPAs on 30 and 15 μm pitch, respectively. Polarimetric imaging with these detectors is demonstrated in this work, showing polarimetric contrast as high as 57±5 % for the QVGA format detectors, while the VGA format arrays provide enhanced spatial resolution at the expense of a lower polarization contrast (≈21±2 %). Finally, to meet the demand of highly sensitive and reliable HD LWIR detectors, a 1280×1024 QWIP detector on 10 μm pitch is the next targeted format of IRnova’s QWIP sensors. In this work, the anticipated performance of such detectors will be presented based on the simulation result using the in-house developed tools. The modelling takes into account the relevant ROIC parameters, as well as scaling of the detector performance with pixel size.

A 1280x1024 Readout Integrated Circuit (ROIC) with 15 μm pixel pitch for MWIR and LWIR applications is designed in 0.18 μm CMOS process. Integrate-Then-Read (ITR) and Integrate-While-Read (IWR) configurations are supported in snapshot mode. The ROIC pixel topology is direct injection (DI) with 3-bit programmable pixel gain. The minimum charge capacity is 0.8 Mé and the maximum full-well-capacity (FWC) is more than 13.2 Mé in ITR mode. 2, 4, 8, and 16 analog video output modes are also supported with a maximum frame rate of 240 fps at 16-output mode. A digital serial interface is used to program timing registers and analog biases. Integration time is programmable with 0.1 μs resolution up to 429 second using 10 MHz clock frequency. The pixel supports binning, windowing and anti-blooming functions. Digital and analog blocks of the ROIC operates using 1.8V and 3.3V supplies. The power consumption is less than 175 mW in 4-output mode of operation at 30 fps.

Digital pixels enable the disassociation of well capacity with noise floor. This allows for better sensitivity for small signals and significantly increased well capacity for large signals. A 640 x 512 format, 20 μm pixel pitch digital pixel readout integrated circuit (DPROIC) for high dynamic range infrared imaging is presented. The architecture uses an extended counting approach that is optimized for low power consumption. This DPROIC boasts a programmable well capacity of 40 Me- in high gain (HG) and >400 Me- in low gain (LG) with no rollover. Read noise is 50 e- rms, HG, and 330 e- rms, LG, for integrate-then-read and 85 e- rms, HG, and 700 e- rms, LG, for integrate-while-read at 80 K. This readout, with LWIR SLS detectors, achieves single digit millikelvin noise-equivalent temperature difference.

Graphene infrared (IR) photodetectors are promising devices that exploit the unique optoelectronic properties of graphene, such as broadband light absorption, rapid response, and high-chemical stability. However, graphene has low absorbance (2.3%), which limits its photo-responsivity. This study investigated the middle-wavelength infrared (MWIR) and long-wavelength IR (LWIR) responsivity enhancements in graphene photodetectors with type-II superlattices (T2SL) employed as photosensitizers. Graphene field-effect transistors (GFETs) were fabricated with the InAs/InGaSb SL photosensitizers on a GaSb substrate. The device was vacuum-cooled and then exposed to light using a filament lamp (wavelength ranges: 3–5 μm and 8–12 μm). The device exhibited MWIR and LWIR photoresponses at 77 K, whereas no photoresponse was observed when photosensitizers were not used. The observed current variations in the photosensitizers and the substrate suggest that photocarriers in the InAs/InGaSb SL modulate the gate voltages on the GFETs, thereby producing the photogating effect. The LWIR photoresponse in conjunction with the photogating effect was enhanced by a factor of 1500 compared with that without the photosensitizers. The results obtained in this study are expected to contribute to the development of high-performance graphene-based IR image sensors.

State-of-the-art mid- and long-wave infrared detectors (IR) are extremely sensitive and are widely used in military and civilian applications. However, these detectors are not able to provide polarization wavelength, directionality, or coherence length of an incident IR source. Our novel IR detector, thermoelectrically coupled nanoantennas (TECNAs), inherently provide these attributes based on their antenna nature. In our talk, we show that the receiving characteristics of TECNAs are determined by their relative position above a cavity etched into the substrate. Our experimental results are supported by simulations that show that TECNAS are capable to detect the directionality of an incident IR beam.

The growing demand for monitoring of specific molecules has created a need for inexpensive and power-efficient light sources. In particular, the mid-infrared (MIR) wavelength range from 3 µm to 6 μm is of high interest for gas-sensing. Aiming at providing reliable broadband and cost-effective alternatives to standard optical gas analysis, nanoplus has developed light-emitting diodes (LEDs) in the MIR. The novel substrate-side-emitting devices rely on the innovative nanoplus technology for distributed feedback (DFB) interband cascade lasers (ICL) and are available at customized wavelengths between 2800 nm and 6500 nm. In this talk we will present various applications which utilize DFB ICLs or MIR LEDs and give a general overview of this technology including latest results from interband cascade photodiodes (PD) and LED-arrays in the MIR.

We built a multispectral data collection system and vehicle testbed for experimentation on vision-based geolocation. The data collection system includes a gimballed mount with VIS, NIR, SWIR, MWIR and LWIR sensors allowing us to compare simultaneous imagery of features or targets across all the atmospheric bands. For geolocation experiments, the testbed is equipped with a dual-GPS inertial measurement unit for true location and orientation, a CAN bus interface to pull vehicle speedometer and odometer data. It is also outfitted with a dual-GPU, rugged, edge computer used to control the system and collect data; the computer is pre-loaded with geospatial data (maps, tower positions, elevation data, etc.) necessary for tracking targets of interest or performing real-time geolocation estimates. 60% of the rear seat was replaced with an electronics rack which also houses a 3kW inverter providing power to all of the equipment. A weather-proof cable pass through was installed in the roof of the truck while a weather-proof enclosure provides wind and rain protection to the roof-mounted equipment. We present multispectral panoramic imagery of flat environments where cellular towers provide ideal references for geolocation and mountainous environments where the landscape and horizon topography provide viable geo-references. We will present an overview of the data collection modes, calibration procedures, and the driving data sets collected to date.

The ability to reliably and accurately ascertain a vehicle’s position is imperative for military operations as well as civilian and commercial navigation systems. Due to the susceptibility of GPS signals to RF spoofing and jamming, alternative means of vehicle self-localization are garnering substantial interest. Vision-based methods are among the most promising in environments with sufficiently distinguishable features such as towers, high-rise structures, and prominent identifiable topographical features. Here, we present a localization approach exploiting multiple spectral bands to identify key prominent scene features and determine vehicle position relative to those features to calculate a global vehicle position and heading. We employ geometric dead-reckoning using visible and LWIR imagery to quantify positional accuracy that is achievable with these bands. We utilize image recognition algorithms to identify features and map these into useful parameters for position extraction, leveraging geospatial data when possible.

Contemporary large-format EO/IR camera systems have several shortcomings as related to 1) their compromise on power or frame rate , 2) their high latency to read out a pixel which impacts their use in tracking applications, and 3) excessive data produced that overload embedded processing. These issues are even more pronounced in space applications which have limited cooling budget, have limited bandwidth, and have limited on-board processing capability. This paper presents a novel low SWAP-C event based ROIC architecture developed to alleviate these issues in large format FPAs. A prototype ROIC developed demonstrates the capability in a 512x512 format with 30um pixel pitch. The ROIC is designed to be stitched up to 4K x 4K array sizes with a 15um pitch. The event based ROIC allows regular frame based imaging, as well as event readout with latencies as low as 20us, while simultaneously reducing system power in hyper-temporal applications by an order of magnitude.

Sunlight reflected on the moving sea surface, often referred as sun glitter, can significantly impact the input data in many application areas like defense, security and border surveillance. The effectiveness of vision-based algorithms like target detection and tracking will decrease remarkably under the sun glitter. To deal with this drawback and give a reaction, the sun glitter should be detected in the scene firstly. In this paper, a straightforward and efficient sun glitter detection and classification method for infrared scenes is introduced. In the first step, the behavior of sun glitter in infrared scenes is analyzed. Two stepped frame buffering strategy with dilation process is used to model the spatiotemporal connection between glitter points. After some thresholding steps, the sun glitter detection decision is taken by checking the total number of glitter candidate pixels. And the classification, which means labeling the sun glitter level as low, mid or high for the scope of this paper, is made by checking the number of blobs that are close to each other. The performance of proposed algorithm is tested with real world infrared dataset, which has 30 different scenarios.

We demonstrate an improvement in the thermal sensitivity of the MelDIR thermal diode infrared sensor which was launched by Mitsubishi Electric Corporation in 2019. To improve the thermal sensitivity of a MelDIR sensor while suppressing any increase in its thermal time constant, we proposed a high thermal absorption pixel structure using an IR absorber composed of a thin metal film. The thin metal film is deposited on the upper layer of the layer of p-n junction diodes. We used a flat structure which has almost the same layer structure as the original MelDIR temperature sensor to confirm the effect of the thin metal film. The measured results showed that the average IR absorbance of the sample at wavelengths from 8 to 14 μm was improved by up to approximately 160%. This value agrees approximately with the calculated results using a Fresnel equation model. Based on these initial results, we numerically estimated the performance of the temperature sensor used in MelDIR using the high thermal absorption pixel structure and verified that the thermal sensitivity of that used in MelDIR had been improved while also suppressing any increase in the thermal time constant.

Black phosphorous (BP) is a promising material for infrared (IR) photodetectors. We theoretically investigated and proposed BP/graphene van der Waals (vdW) heterostructure-based metasurface wavelength-selective IR photodetectors that operate on the basis of plasmonic resonance. The proposed structure consists of a top layer consisting of periodic monolayers of BP, a graphene-based vdW heterostructure layer on a middle dielectric layer, and a bottom reflector. The periodic BP produces polarization-dependent localized surface plasmon resonance (LSPR) according to its armchair or zigzag edge, resulting in strong wavelength-selective absorption. The graphene layer can enhance the LSPR of BP, inducing propagating surface plasmon resonance on graphene. These results will contribute to the realization high-performance BP/graphene-based IR photodetectors.

Uncooled bolometer-type infrared detector's performance is strongly limited by temperature coefficient of resistance (TCR) of the resistor. Single-walled carbon nanotubes (SWCNTs) are expected as promising materials with high TCR, and high chemical stability. In a preliminary experiment TCR of semiconducting CNT networks extracted by using non-ionic surfactants was confirmed to be -6%/K. Electric properties of SWCNT networks were strongly affected by its morphology. Optimization of SWCNT network morphology is important to maximize the detectability of SWCNT bolometers.

Graphene/InSb heterojunction mid-infrared photodiodes for infrared image sensors exhibit high responsivity, low noise, and excellent detection performance. However, the performance of each pixel varies owing to the instability of the graphene/InSb interface state. In this study, the performance variation was addressed by inserting an interfacial layer at the graphene/InSb interface. A few-nanometer-thick HfO2 interfacial layer was inserted at the graphene/InSb interface. Compared to devices without an interfacial layer, those with an interfacial layer exhibited greatly improved minimum noise equivalent temperature difference (NETD) and pixel-by-pixel NETD variation. This was due to the stabilization of the InSb surface and increased photocurrent caused by photoswitching, which significantly changed the Fermi level of the graphene. The insertion of the HfO₂ interfacial layer improved the standard deviation to less than 1/30. Thus, inserting an interfacial layer at the graphene/InSb interface is expected to facilitate the development of high-performance infrared image sensors with low variability.

Systems using low-cost infrared (IR) sensors have gained more and more interest in recent years, such as smart video surveillance, driver assistance system, robotics, and military applications. To meet the total cost limit of these systems, low resource processers are desired. In this paper, we propose a scalable resource-driven pedestrian detection approach for low-cost IR sensors. The proposed method introduces a feature-importance-sampling based feature selection for the IR sensors which enables scalable selection of the number of effective features according to the resource requirements. To obtain high performance detection with the few selected effective features and hard-to-obtain data of low-cost IR sensors, we adopt transfer learning from easy-to-obtain visible images to train the pedestrian detector. In this way, optimal IR system design can be determined based on the trade-offs between performance and resources. Using HOG and SVM based pedestrian detection methods, we evaluate the proposed method on Thermal Diode lnfraRed (TDIR) sensor and find that key-points of the human body that observed in the TDIR images, such as the contours of the body, are selected as the most crucial attributes. As a result of training these selected features on visible images, one can reduce memory use by 1/30 and increase detection speed by about 1/6 while maintaining competitive detection performance. To summarize, the proposed work presents a scalable resource-driven method for pedestrian detection for low-cost IR sensors with high accuracy and low resource consumption.

For the next-generation 8 to 14μm long-wavelength infrared (LWIR) sensing, type-II superlattices (T2SLs) detectors have enormous potential to appeal to various applications including space, medical imaging, and defense. In typical absorber design, the sufficiently thick active layer (AL) is required to achieve high quantum efficiency (QE), whereas it can cause a high dark current and increase the cost. Moreover, a simple increase in AL thickness does not provide an increase in QE due to the limited carrier lifetime in T2SLs. A possible solution to the weak absorption in the AL of the T2SL-based detectors involves incorporating the AL into an optically resonant cavity. In this work, optically enhanced QE for the broadband T2SL nBn detectors will be presented through the guided-mode-resonance (GMR) structures on the top surface of the T2SL. We fabricated T2SL nBn detectors with periodic gratings on the top surface. The devices showed much-enhanced QE due to multiple resonances, as well as Fabry-Perot resonance in the thin AL with a lower dark current characteristic than the reference T2SL detector. Furthermore, we also found the broadening of the cutoff wavelength, which is typically limited by the material property, by scaling the dimension of the diffraction grating for a strong resonance beyond the cut-off region. In conclusion, the GMR-based LWIR T2SL detectors can show a significant performance enhancement in QE and extend the detection range beyond the cut-off wavelength while maintaining a low dark current.

A large format, high density integration, high-performance infrared detectors are used in a wide range of imaging applications. However, it is difficult to fabricate a hybrid chip of a high performance infrared detector because of the low flip-chip bonding and under-fill process yields. In this work, the large format hybrid chip fabrication process with wafer level integration schemes is presented. In particular, the structure of the fabricated hybrid chip and the method of fabrication a hybrid chip by Au-to-Au bonding. Finally, the result of a hybrid chip made of pixel pitch 20μm is presented.

We investigate the origin of dark current by studying the bulk and surface recombination mechanisms using temperaturedependent minority carrier lifetime measurements of GaAsSb p-i-n devices. We measure the equilibrium carrier concentration (background doping concentration) by transient microwave reflectance and compare with capacitancevoltage measurements. Room temperature minority carrier lifetime results are presented, as well as initial low temperature measurements.

The issue of climate change is of ever-increasing concern to the global community and key to understanding the problem is the characterization of sources, sinks and transportation of greenhouse gases (GHGs) such as atmospheric carbondioxide (CO₂). This requires space-based observations to provide a truly global picture, due to the sparse nature of terrestrial monitoring sites. The sparseness of these sites, especially along the tropics makes it difficult to determine the extent of carbon fluxes around the equatorial regions. To fully characterize the problem, a high spatial and temporal resolution is needed. Although atmospheric CO₂ can be detected optically by small satellites, precise quantitative mapping requires extremely high precision (0.3% to 0.5%) measurements of gas concentrations at spectral absorption bands of 1.56-1.62μm and 1.92-2.06μm wavelengths, using high resolution spectrometers (e.g. 0.27 cm^-1 resolution at a signal-to-noise ratio (SNR) of >300:1). This normally requires high-performance, large and complex instruments, carried by single, large and expensive satellites. No single satellite can provide high temporal resolution. Therefore, a constellation of satellites is needed to resolve the underestimation of CO₂ uptake (sink) by the terrestrial biosphere. In this paper, we present the developmental progress of SHACS—a low-cost, compact, precision, robust, no-moving part spatial heterodyne atmospheric carbon-dioxide spectrometer, and also present some results of the Total Column Carbon-dioxide (TCCO₂) measurements obtained at Surrey in June 2019. The SHACS instrument has achieved a high measurement precision of <0.3% and a high spectral resolution of 0.22 cm^-1 at a high SNR of >900:1.

There is an established need for a readily available, low cost, low noise, high performance ROIC and FPA to support rapid turnaround low-cost New Space earth observing missions. However, there is a dearth of space qualified commercially available sensors. The design presented is developed to address a variety of push broom TDI scanner format used in earth observing payload. Both panchromatic and multispectral sensors are developed. The sensor is highly configurable for the mission with programmable gain, on-chip programmable bit depth ADC, on-chip programmable biases, programmable digital TDI stages, and serial digital output interface. The ROIC offers embedded silicon detectors, that can be optimized for producing monolithic FPAs for UV, for visible band and for NIR band. The design is also compatible with a variety of IR detectors from SWIR through MWIR bands. This ROIC will enable a cheaper FPA alternative for rapid space missions.