This work focuses on a bimodal vision system which was previously demonstrated as a relevant sensing candidate for detecting and tracking fast objects by combining the unique event-based sensor features i.e. high temporal resolution, reduced bandwidth needs, low energy consumption, and passive detection capabilities with the high-spatial resolution of a RGB camera. In this study, we aim to propose a model based on the principle of attentional vision for real-time detection and tracking of UAVs, taking into account computing and on-board resource constraints. A laboratory demonstrator have been proposed to evaluate the operational limits in terms of computation time, system performances (including target detection) versus speed. Our first indoor and outdoor tests revealed the interest and potential of our system to quickly detect objects flying at hundreds of kilometers an hour.

With the improvements of camouflage capacities, some threats are more and more difficult to detect with conventional imaging devices. For example, some camouflage techniques allow being undetectable both on color camera and thermal camera, at the same time.

By implementing a mosaic multispectral technology, Safran designed a compact and lightweight camera able to create a snapshot multispectral image of nine spectral bands. The multispectral images acquired with this camera are numerically treated and projected in an adapted space in order to maximize the visual contrast between each point and its surrounding.

Last year, Safran designed a real-time camouflage detection device integrated in an armored vehicle sight system and conducted several test campaigns, mainly in Mid-European context. This work was performed in the context of a collaborative project between ONERA and Safran, with the support of DGA.

With this first campaign, we compared conventional cameras and a multispectral camera with different criteria such as metric comparison and visual comparison. This study helped us on the optimization of our computer algorithm in order to maximize our detection rates.

High-performance visual detection and tracking of small unmanned aerial vehicles (UAVs) at a distance is crucial for laser-based countermeasures, which require precise and accurate aiming to be effective. Achieving robust and accurate real-time tracking while dealing with dynamic non-cooperative targets and challenging lighting conditions poses significant difficulties for tracking solutions. Event cameras represent a new perception paradigm to overcome these hurdles by registering changes in illumination asynchronously with low latency, high temporal resolution and large dynamic range. The cameras yield a sparse stream of event information which encodes changes in the scene. Accumulating events into frames and using traditional computer vision techniques can be applied, but sacrifice some of the information contained in the event stream. Asynchronous methods have shown promise in utilizing the event stream more efficiently, but are not yet at the same technical maturity as frame-based alternatives. This work introduces and evaluates two event-based end-to-end detection and tracking systems suitable for real-time applications. One asynchronous and one accumulating method were studied. The different methods are evaluated using an annotated event-dataset containing recordings of multi-rotor UAVs operating in an outdoor setting to compare the performance.

Single photon avalanche diode (SPAD) sensors can detect the occurrence of a single photon and record binary intensities, count rates or event timing with virtually no noise. This revolutionary new optical sensing concept has the potential to trigger a paradigm shift that will change the way we talk and think about optical sensing. In recent years, this technology has made significant advances, and the latest sensors offer high-resolution arrays with high temporal and spatial resolution.

In this paper, we present results achieved at ISL that demonstrate how single-photon imaging combined with computational methods differs from classical imaging methods. We show how we can extract and reconstruct new, previously unattainable information from scenes.

ISL has investigated passive single photon counting to reconstruct the photon flux imaging the sensor array. We could reconstruct image information and obtained up-scaling by application of convolutional neural networks, reduced noise and motion blur by computer vision algorithms. Finally, we extracted modulation frequencies by Fourier analysis and obtained event-based neuromorphic imaging.

Further, we have studied laser-based active imaging of single photons to measure the round-trip path length of light pulses for ranging and 3D imaging. We have analyzed multi-bounce photon path to estimate the size of cavities and to improve vision through scattering media such as dense fog. Finally, we investigated SPAD sensing for the reconstruction of objects outside the direct line of sight in non-line of sight (NLOS) sensing approaches.

We demonstrated a transceiver system for underwater three-dimensional imaging, based on a 64 × 32 macro-pixel direct time-of-flight (dToF) SPAD detector array fabricated using complementary metal-oxide-semiconductor (CMOS) technology. The sensor featured integrated multi-event time-to-digital converters (METDC) per macro-pixel, and operated in a time-gated mode which allowed the improved rejection of the backscattering from the signal return. The performance of the system was assessed in a controlled underwater environment, utilizing a moving target placed in a water tank at a stand-off distance of 1.45 m in water. The system demonstrated rapid 3D imaging by using short acquisition times of 1 ms, which corresponds to a frame rate of 1000 fps. Depth and intensity profiles were obtained at attenuation levels equivalent up to 5.5 attenuation lengths between the transceiver and the target, and using average optical power of up to 32 mW.

This study introduces an affordable approach to developing a millimeter-wave (MMW) focal plane array (FPA) using glow discharge detectors (GDDs) with rapid readout capabilities. The system investigates the effect of MMW incidence on GDD discharge current and employs a scanning mechanism with a step motor for sub-pixel imaging. The experimental setup features an 8x8 FPA with 64 GDD detectors, MMW source, optical components, and a data acquisition platform. Leveraging electrical detection and a signal processing algorithm, the system captures low-intensity MMW signals with each GDD value corresponding to a pixel in the image. The Fourier transform (FFT) is used to extract information from the modulated signals. The FPA demonstrated an improved imaging duration, reducing acquisition time by 75% compared to prior studies, with an 8x8 image captured in just 4 seconds. Responsivity was measured at 21.8 V/W, detecting signals as weak as 0.95 nW. These advancements offer significant benefits for time-sensitive applications like security and dynamic object tracking. The study proposes enhancing resolution through larger arrays and integrating image processing with AI techniques to further improve performance. This research provides a cost-effective solution for MMW imaging, addressing existing detector limitations.

This paper presents an optimized, reconfigurable, ultra-compact mode converter device designed for efficient conversion between transverse electric (TE) modes, specifically TE₀ and TE₁. The mode conversion is achieved through the manipulation of an antimony selenide (Sb₂Se₃) phase-change material (PCM) layer, integrated on the top surface of a silicon waveguide. The device’s performance was investigated numerically at a wavelength of 1550 nm, focusing on optimizing the size of the phase-change layer to enhance transmission and minimize losses. With a footprint of just 4 μm × 3 μm, this converter is highly suitable for integrated photonic systems, including optical communication, signal processing, and photonic circuits. The dynamic reconfigurability is facilitated by the reversible phase transitions of Sb₂Se₃ between its amorphous and crystalline states, enabling low-loss mode conversion. By leveraging this material and its properties, the converter achieves high transmission efficiency with minimal insertion loss. These findings pave the way for more compact and efficient photonic devices, addressing the critical demand for reconfigurable mode converters in integrated photonics. This work represents a significant advance in device performance and footprint reduction, positioning it as a state-of-the-art solution for next-generation photonic applications.

This paper explores the modelling of an integrated photonic power divider that utilizes the degree of crystallization of an Sb₂Se₃ phase-change material (PCM) patch to achieve variable power distribution ratios. The fraction of crystallinity of Sb₂Se₃, ranging from 0 to 1, provides a dynamic control mechanism for the device. Simulations were performed using a combination of optimization algorithms and wave optics interface within COMSOL Multiphysics^® to arrive at the optimum distribution of amorphous and crystalline fractions to achieve a required power distribution. The transmission coefficients serve as control variables to direct light to the upper or lower port, allowing the power divider to function as a reconfigurable and compact photonic device. These findings provide valuable insights into the use of PCMs for advancing photonic power dividers.

Continuous capture microwave image detection and ranging (m-Widar) is an emerging sensing concept which combines sparse image reconstruction with an inverse light transport model. This enables a ‘single pixel microwave camera’ to densely image a surrounding environment in milliseconds via reflected microwaves, while requiring minimal hardware and set up. As such, m-Widar has many potential novel applications, including sensing through walls, vehicle detection, and standoff vital-sign monitoring. This work describes m-Widar sensing fundamentals and the challenges involved for practical multi-object detection and tracking. We present a Bayesian Hidden Markov Model (HMM) tracking approach to address these issues, along with proof-of-concept simulation assessments and directions for ongoing work.

Time-correlated single-photon lidar is a technology that can provide high resolution lidar measurements, with low laser power, providing many applications in defence and security. One problem with the InGaAs/InP single photon avalanche detectors (SPAD), which are sensitive around 1.55 μm, is that they suffer from afterpulsing from trapped carriers. This is usually solved by having a hold-off time before reapplying the voltage to the detector, after a detection, to let the trapped carriers dissipate. For InGaAs SPAD this hold-off time, during which the detector is insensitive to photons, needs to be several microseconds to avoid runaway afterpulsing and tens of microseconds to get low afterpulsing.

This paper describes a model to calculate the received number of laser photons in lidar experiments, under the influence of detector dark counts, ambient background, afterpulsing and hold-off time. By using characterization data of photon detection efficiency, dark count rate and afterpulsing, for different detector settings, graphs of optimum operation points for different signal and background levels are simulated. Some comparisons to experimental data are performed.

We present numerical studies on InSb/InAs_1-xSb_x type-II superlattices (T2SL) for applications in multi-spectral-infrared detectors spanning peak detection wavelengths from 2.3 to 5.5 μm. The wavelength tuneability is achieved by varying the mole fraction x of the ternary compound indium arsenide antimonide (InAs_1-xSb_x) from 0.05 to 0.95. The simulated T2SL comprises 50 periods of 1.4 nm of InAs_1-xSb_x and 2.8 nm of indium antimonide (InSb). The use of T2SL structures allows to achieve bandgap tuneability, higher quantum confinement, lower noise and faster detection, leading to higher temperatures of operation compared to their bulk semiconductor counterpart, with similar performance. A T2SL p-i-n photodetector stack was simulated using software NextNano, simulating optical absorption with the energy bandgap of the T2SL tuned from 0.53 eV (2.3 μm in wavelength) to 0.22 eV (5.5 μm). The bandgap tunability of our simulated T2SL detector makes it possible to match the peak detection wavelengths to the absorption lines of several greenhouse gases such as methane and carbon dioxide. The simulated layer stack implements a multi-spectral detector suitable for epitaxial growth on gallium arsenide (GaAs) hence integration with GaAs-based readout electronics, enabling a monolithic multispectral imaging array technology, with applications in defence/security healthcare, and environmental monitoring.

Mid-wave Infrared (MWIR) detectors based on Indium Antimonide (InSb) and Mercury Cadmium Telluride (MCT) are useful for a range of applications in autonomous systems, industrial and environmental monitoring, and night vision for security and defence. Their requirement for cryogenic cooling, however, limits their applications, decreasing their portability, increasing power consumption, and slowing down startup times. InSb detectors operated at room temperature would be more portable, consume less power, and speed up startup times. However, their limited sensitivity, high noise, and longer response times make them unsuitable for high-end applications. Plasmonic metasurfaces have been widely deployed for optical absorption enhancement, but they have a narrowband response. In this work, we present a numerical study that investigates the feasibility of using transparent conducting oxide (TCO) as the epsilon-near-zero (ENZ) layer surmounted by a gold plasmonic metasurface as the top electrode. This Metal-Semiconductor-Metal Schottky devices based on InSb shows near-unity broadband absorption of incident MWIR light. The synergy between ENZ layer and the designed plasmonic metasurface allows for a homogeneous broadband absorption in the underlying InSb layer, insensitive to the polarisation and angle of incident light. The thickness of the TCO layer, carrier density, and the parameters of the plasmonic metasurface are optimised to produce high absorption (>90%) in the wavelength span 3.37–4.55 μm, corresponding to a high atmospheric transmission channel. The inclusion of a TCO layer enhances the optical performance while also providing an electrical path for the photo-generated current. The ENZ-enabled plasmonic meta-absorber device that we propose has therefore the potential to achieve higher detectivity, paving the way for the development of nextgeneration miniaturized, high-speed, uncooled MWIR focal plane arrays (FPAs).