Target Resolution Research Articles

Near-Field IPO for Analysis of EM Scattering from Multiple Hybrid Dielectric and Conductor Target and High Resolution Range Profiles

Aiming at improving the accuracy and efficiency of scattering information from multiple targets in near-field regions, this paper proposes a near-field iterative physical optics (IPO) method based on a modified near-field Green’s function for the composite electromagnetic scattering analysis of multiple hybrid dielectric and conductor targets. According to the electric field and magnetic field integral equation, the electric and magnetic current were updated utilizing the Jacobi iteration method. Then, by introducing an expansion center lying in the neighborhood of the source point, Green’s function was modified for near-field scattering between multiple hybrid dielectric and conductor targets. To accelerate the implementation of the procedure, the multilevel fast multipole method, the fast far-field approximation, and parallel multicore programming were introduced. Numerical results indicate that there is good agreement between the results calculated by the near-field IPO method and MLFMM solver in commercial software FEKO while significantly reducing the computational burden. To fully exploit the scattering information, the high resolution range profiles (HRRP) of different targets under different conditions were analyzed, which can be further applied for automatic target detection and recognition.

Soil moisture estimates at 1 km resolution making a synergistic use of Sentinel data

Abstract. Very high-resolution (∼10–100 m) surface soil moisture (SM) observations are important for applications in agriculture, among other purposes. This is the original goal of the S2MP (Sentinel-1/Sentinel-2-Derived Soil Moisture Product) algorithm, which was designed to retrieve surface SM at the agricultural plot scale by simultaneously using Sentinel-1 (S1) backscatter coefficients and Sentinel-2 (S2) NDVI (Normalized Difference Vegetation Index) as inputs to a neural network trained with Water Cloud Model simulations. However, for many applications, including hydrology and climate impact assessment at regional level, large maps with a high resolution (HR) of around 1 km are already a significant improvement with respect to most of the publicly available SM datasets, which have resolutions of about 25 km. In this study, the S2MP algorithm was adapted to work at 1 km resolution and extended from croplands to herbaceous vegetation types. A target resolution of 1 km also allows the evaluation of the interest in using NDVI derived from Sentinel-3 (S3) instead of S2. Two sets of SM maps at 1 km resolution were produced with S2MP over six regions of ∼104 km2 in Spain, Tunisia, North America, Australia, and the southwest and southeast regions of France for the whole year of 2019. The first set was derived from the combination of S1 and S2 data (S1 + S2 maps), while the second one was derived from the combination of S1 and S3 (S1 + S3 maps). S1 + S2 and S1 + S3 SM maps were compared to each other, to those of the 1 km resolution Copernicus Global Land Service (CGLS) SM and Soil Water Index (SWI) datasets, and to those of the Soil Moisture Active Passive (SMAP) + S1 product. The S2MP S1 + S2 and S1 + S3 SM maps are in very good agreement in terms of correlation (R≥0.9), bias (≤0.04 m3 m−3), and standard deviation of the difference (SDD≤0.03 m3 m−3) over the six domains investigated in this study. In a second step, the S1 + S3 S2MP maps were compared to the other HR maps. S1 + S3 SM maps are well correlated to the CGLS SM maps (R∼0.7–0.8), but the correlations with respect to the other HR maps (CGLS SWI and SMAP + S1) drop significantly over many areas of the six domains investigated in this study. The highest correlations between the HR maps were found over croplands and when the 1 km pixels have a very homogeneous land cover. The bias among the different maps was found to be significant over some areas of the six domains, reaching values of ±0.1 m3 m−3. The S1 + S3 maps show a lower SDD with respect to CGLS maps (≤0.06 m3 m−3) than with respect to the SMAP + S1 maps (≤0.1 m3 m−3) for all the six domains. Finally, all the HR datasets (S1 + S2, S1 + S3, CGLS, and SMAP + S1) were also compared to in situ measurements from five networks across five countries, along with coarse-resolution (CR) SM products from SMAP, SMOS, and the European Space Agency Climate Change Initiative (CCI). While all the CR and HR products show different bias and SDD, the HR products show lower correlations than the CR ones with respect to in situ measurements. The discrepancies in between the different HR datasets, except for the more simple land cover conditions (homogeneous pixels with croplands) and the lower performances with respect to in situ measurement than coarse-resolution datasets, show the remaining challenges for large-scale HR SM mapping.

A fast grid generation algorithm for local irregular parts of hexagonal discrete global grid systems

ABSTRACT Discrete Global Grid Systems (DGGS) provide a multi-resolution discrete representation of the Earth and are preferable for the organization, integration, and analysis of large and multi-source geospatial datasets. Generating grids for the area of interest is usually the premise and basis for DGGS applications. Owing to incongruent hierarchies that restrict the multi-resolution applications of hexagonal DGGS, current grid generation of hexagonal DGGS for local areas mainly depends on inefficient single-resolution traversal methods by judging the spatial relationship between each cell and the area. This study designs a fast generation algorithm for local parts of hexagonal DGGS based on the hierarchical properties of DGGS. A partition structure at intervals of multiple levels is first designed to ensure the coverage relevance between parent and children cells of different levels. Based on this structure, the algorithm begins with coarser resolution grids and recursively decomposes them into the target resolution, with multiple decomposition patterns used and a unique condition proposed to make the generated grids without gaps or overlaps. Efficient integer coordinate operations are used to generate the vast majority of cells. Experimental results show that the proposed algorithm achieves a significant improvement in efficiency. In the aperture 4 hexagonal DGGS, the efficiency ratio of the proposed and traversal algorithms increases from six times in level 14 to approximately 339 times in level 18. This study provides a solid foundation for subsequent data quantization and multi-resolution applications in hexagonal DGGS and has broad prospects.

Weak Underwater Acoustic Target Detection and Enhancement with BM-SEED Algorithm

Bearing time record (BTR) is widely used in the field of passive sonar information processing for target detecting and tracking. One of its challenges is to obtain high-resolution beamforming power spectral to facilitate the directions estimating of arrivals, namely DOA estimating. This paper proposes a new algorithm framework Block Matching - Subband Extrema Energy Detection (BM-SEED) to supplement the resolution and enhance the weak targets of BTRs. It extracts the peak features of target trajectories within the conventional beamforming (CBF)-based BTRs by using an improved sub-band peak energy detection (SPED) algorithm, and then enhances the weak targets and suppresses the false-alarm targets through time-spatial autocorrelation analysis. Simulations and sea-trial data evaluations demonstrate that the method of this paper can enhance weak targets, eliminate the interferences of false targets and improve the resolution of BTR targets to 1 degree under −20 dB.

Patchy Kinetic Sunyaev–Zel’dovich Effect with Controlled Reionization History and Morphology

Using the novel semi-numerical code for reionization AMBER, we model the patchy kinetic Sunyaev–Zel’dovich (kSZ) effect by directly specifying the reionization history with the redshift midpoint z mid, duration Δz, and asymmetry A z. We further control the ionizing sources and radiation through the minimum halo mass M h and the radiation mean free path λ mfp. AMBER reproduces the free-electron number density and the patchy kSZ power spectrum of radiation–hydrodynamic simulations at the target resolution (1 Mpc h −1) with matched reionization parameters. With a suite of (2 Gpc/h)3 simulations using AMBER, we first constrain the redshift midpoint 6.0 < z mid < 8.9 using the Planck 2018 Thomson optical depth result (95% CL). Then, assuming z mid = 8, we find that the amplitude of scales linearly with the duration of reionization Δ z and is consistent with the 1σ upper limit from South Pole Telescope (SPT) results up to Δ z < 5.1 (Δ z encloses 5%–95% ionization). Moreover, a shorter λ mfp can lead to a ∼10% lower and a flatter slope in the scaling relation, thereby affecting the constraints on Δ z at ℓ = 3000. Allowing z mid and λ mfp to vary simultaneously, we get spectra consistent with the SPT result (95% CL) up to Δ z = 12.8 (but A z > 8 is needed to ensure the end of reionization before z = 5.5). We show that constraints on the asymmetry require ∼0.1 μ k 2 measurement accuracy at multipoles other than ℓ = 3000. Finally, we find that the amplitude and shape of the kSZ spectrum are only weakly sensitive to M h under a fixed reionization history and radiation mean free path.

Coherent Multi-Dwell Processing of Un-Synchronized Dwells for High Velocity Estimation and Super-Resolution in Radar

This paper describes a coherent multi-dwell processing (CMDP) method for high velocity estimation and super-resolution in search and track, while search (TWS) radar modes use an un-conventional signal processing algorithm that exploits multi-dwell transmissions. The existence of the multi-dwell waveform is necessary for visibility needs by un-folding the target’s velocity and range ambiguity and is proposed to be utilized for high velocity estimation and super-resolution. In this paper, the proposed scheme is shown to result in improved velocity estimation and doppler resolution performance for un-ambiguous targets in comparison to classical radar processing. The processing concept uses the same transmitted waveform (WF) and time duration without the need to increase the time on target (TOT) through sophisticated coherent concatenation of the received dwells with velocity compensation between the dwells. The phase compensation in receive mode is implemented for each target according to its characteristics, which means that target velocities are estimated in each dwell separately. The notable result of the CMDP is the linear doppler resolution improvement obtained with the given search resources and without knowing the target characteristics in advance or the dwell delay time. Other possible benefits of this process are the ability to achieve larger detection ranges and high-angle measurement precisions in search mode due to the higher signal-to-noise ratio (SNR) of the extended dwell and the ability to track more targets due to efficient time and resource management. An outstanding opportunity to exploit the CMDP is by combining missions in phased array (PA) radars, meeting the multi-objective needs of both high spatial scan rates for illuminating the target and high doppler estimation and resolution performance.

Mapping Brain Motor Functions Using Transcranial Magnetic Stimulation with a Volume Conductor Model and Electrophysiological Experiments.

Transcranial magnetic stimulation (TMS) activates brain cells in a noninvasive manner and can be used for mapping brain motor functions. However, the complexity of the brain anatomy prevents the determination of the exact location of the stimulated sites, resulting in the limitation of the spatial resolution of multiple targets. The aim of this study is to map two neighboring muscles in cortical motor areas accurately and quickly. Multiple stimuli were applied to the subject using a TMS stimulator to measure the motor-evoked potentials (MEPs) in the corresponding muscles. For each stimulation condition (coil location and angle), the induced electric field (EF) in the brain was computed using a volume conductor model for an individualized head model of the subject constructed from magnetic resonance images. A post-processing method was implemented to determine a TMS hotspot using EF corresponding to multiple stimuli, considering the amplitude of the measured MEPs. The dependence of the computationally estimated hotspot distribution on two target muscles was evaluated (n = 11). The center of gravity of the first dorsal interosseous cortical representation was lateral to the abductor digiti minimi by a minimum of 2 mm. The localizations were consistent with the putative sites obtained from previous EF-based studies and fMRI studies. The simultaneous cortical mapping of two finger muscles was achieved with only several stimuli, which is one or two orders of magnitude smaller than that in previous studies. Our proposal would be useful in the preoperative mapping of motor or speech areas to plan brain surgery interventions.

On the Use of Native Resolution Backscatter Intensity Data for Optimal Soil Moisture Retrieval

The accuracy of soil moisture estimated from Synthetic Aperture Radar backscatter data at high resolution is limited by speckle. Common practice to mitigate speckle is to multilook the data prior to retrieving soil moisture. While multilooking indeed reduces speckle, it also decreases the spatial resolution and removes possibly useful high resolution information from the data. We therefore hypothesised that using higher resolution backscatter data for soil moisture retrieval would lead to higher retrieval accuracies. A high-resolution field study combined with a synthetic experiment showed that calculating soil moisture prior to multilooking to the final target resolution (calculate-then-average, CtA) has substantial advantages over the average-then-calculate (AtC) approach. Currently, the AtC strategy is most often applied in soil moisture studies, mainly due to its computational advantage compared to the CtA approach. We show that by making use of a higher source resolution backscatter data than the target resolution, we could improve the soil moisture retrieval over an agricultural field.

Influence of UAV Jittering on Sensing Performance in ISAC System

The integrated sensing and communication (ISAC) based on unmanned aerial vehicle (UAV) is considered to be an important candidate technology for future wireless communication, which can overcome the problem that the line-of-sight (LoS) path of ground communication in millimeter wave or terahertz band is easily blocked. However, recent studies have shown that due to the change of wind speed, the vibration of wing and other factors, UAV will appear jittering effect, which will seriously affect the communication performance. With the continuous research of ISAC, it is also necessary to explore the impact of UAV jittering on the sensing performance. This paper first proposes the UAV jittering model, then takes the target angle sensing scheme based on time-division wave scanning as an example, and explores the influence of UAV jittering on target sensing. This paper finds that UAV jittering has great influence on the accuracy and resolution of target sensing. Simulation results confirm this conclusion.

Sound Speed Estimation for Distributed Aberration Correction in Laterally Varying Media.

Spatial variation in sound speed causes aberration in medical ultrasound imaging. Although our previous work has examined aberration correction in the presence of a spatially varying sound speed, practical implementations were limited to layered media due to the sound speed estimation process involved. Unfortunately, most models of layered media do not capture the lateral variations in sound speed that have the greatest aberrative effect on the image. Building upon a Fourier split-step migration technique from geophysics, this work introduces an iterative sound speed estimation and distributed aberration correction technique that can model and correct for aberrations resulting from laterally varying media. We first characterize our approach in simulations where the scattering in the media is known a-priori. Phantom and in-vivo experiments further demonstrate the capabilities of the iterative correction technique. As a result of the iterative correction scheme, point target resolution improves by up to a factor of 4 and lesion contrast improves by up to 10.0 dB in the phantom experiments presented.

Research on Ultra-Wideband NLFM Waveform Synthesis and Grating Lobe Suppression.

Ultra-wideband (UWB) nonlinear frequency modulation (NLFM) waveforms have the advantages of low sidelobes and high resolution. By extending the frequency domain wideband synthesis method to the NLFM waveform, the synthetic bandwidth will be limited, and the grating lobe will grow as the number of subpulses increases at a fixed synthetic bandwidth. Aiming for the highly periodic grating lobes caused by equally spaced splicing and small subpulse time-bandwidth products (TxBW), a multisubpulse UWB NLFM waveform synthesis method is proposed in this paper. Random frequency hopping and spectral correction are utilized to disperse the energy of periodic grating lobes and optimize the matched filter of the subpulse, thereby reducing notches and Fresnel ripples in the synthesized spectrum. The results of the hardware-in-the-loop simulation experiment show that the peak sidelobe ratio (PSLR) and the integral sidelobe ratio (ISLR) of the NLFM synthetic wideband waveform (SWW) obtained by 50 subpulses with a bandwidth of 36 MHz are improved by 4.8 dBs and 4.5 dBs, respectively, when compared to the frequency domain wideband synthesis method, and that the grating lobe is suppressed by an average of 10.6 dBs. It also performs well in terms of point target resolution, and it has potential for 2D radar super-resolution imaging.

Focus image scanning microscopy for sharp and gentle super-resolved microscopy

To date, the feasibility of super-resolution microscopy for imaging live and thick samples is still limited. Stimulated emission depletion (STED) microscopy requires high-intensity illumination to achieve sub-diffraction resolution, potentially introducing photodamage to live specimens. Moreover, the out-of-focus background may degrade the signal stemming from the focal plane. Here, we propose a new method to mitigate these limitations without drawbacks. First, we enhance a STED microscope with a detector array, enabling image scanning microscopy (ISM). Therefore, we implement STED-ISM, a method that exploits the working principle of ISM to reduce the depletion intensity and achieve a target resolution. Later, we develop Focus-ISM, a strategy to improve the optical sectioning and remove the background of any ISM-based imaging technique, with or without a STED beam. The proposed approach requires minimal architectural changes to a conventional microscope but provides substantial advantages for live and thick sample imaging.

Automatic grid refinement for thin material layer etching in process TCAD simulations

Thin material layers are common structures in modern semiconductor device fabrication and are particularly necessary for light-emitting diodes and three-dimensional NAND memory devices. Such layers are not only deposited on the flat wafer surface but are also partially removed during subsequent etching steps. Level-set based process TCAD simulations are capable of representing flat thin material layers, such as those occurring after deposition, with sub-grid accuracy. However, topographical changes during etching processes modeled via Boolean operations expose the low underlying grid resolution, leading to detrimental artifacts. We present a novel algorithm that analyzes the thickness of all material layers and derives a refined target resolution for local regions of thin layers affected by the etching process. This allows to accurately represent topographical changes in thin layers by hierarchically refining the grid without unnecessary refinement in unaffected regions of the domain. We simulate the fabrication of a light-emitting diode device using our algorithm to automatically predict the optimal resolution for all etched material layers. Our algorithm selects efficient refinement factors to obtain the local target resolutions of the hierarchical grids, and achieves a three times faster computation time than a benchmark refinement algorithm based on topographical features.

The Importance of Very-High-Resolution Imagery to Map Invasive Plant Species: Evidence from Galapagos

Invasive species are one of the main threats to biodiversity worldwide, and the Galapagos Islands are no exception. With the need to control many invasive plant species, accurate distribution maps of invasive plant species are crucial for cost-effective management actions. To guide the selection of appropriate multispectral satellite imagery for this, we evaluated the effects that spatial resolution has on the mapping accuracy of the most invasive plant species in Galapagos with different “growth forms”: (1) tall tree: Cuban cedar (Cedrela odorata), (2) medium tree: guava (Psidium guajava), and (3) shrub: blackberry (Rubus niveus). We developed a mapping methodology based on very high resolution (VHR, WorldView-2) imagery and visual interpretation from orthophotos obtained from unmanned aerial vehicles for training and validation. We then compared our VHR mapping results with medium resolution (MR, Landsat) mapping results and calculated the overall accuracy (OA) and Kappa from confusion matrices for each target species and resolution based on the visual interpretation of Google Earth imagery. The results showed that the OA of the maps produced with VHR was significantly higher than the ones produced with MR. The OA was higher for the tall tree growth form, followed by the shrub and the medium tree growth form. Kappa estimates of &lt;0.5 for MR for the medium tree and shrub growth forms demonstrated its unsuitability for accurate mapping. While MR may be sufficient for mapping the tall tree growth form, we concluded that VHR is indispensable for mapping the medium tree and shrub growth forms.

Applying Deep Learning Cancer Subtyping Algorithms Trained on Physical Slides to Multiphoton Imaging of Unembedded Samples

Abstract Introduction/Objective Deep learning algorithms on digital images of physical tissue slides have shown potential improvements in accuracy and precision of diagnostic interpretation of neoplastic histology. Clustering-constrained- attention multiple-instance learning (CLAM) is one such method that identifies diagnostic sub-regions to accurately classify whole slides. Often, algorithm performance degrades when deployed on datasets that differ from the original set and it is subject to physical slide preparation variability. New multiphoton imaging modalities have potential workflow and quality advantages over physical slides, producing images analogous to whole slide imaging (WSI) without cutting artifacts, but performance of existing algorithms trained on digitized physical slides and applied to multiphoton images remains completely unknown. Given this, we aimed to test the performance of CLAM algorithms for subtyping renal cell carcinoma (RCC) and lung cancer (LC) applied to pseudo-colored multiphoton WSI. Methods/Case Report Clinical RCC and LC surgical resection samples were processed and imaged by Clearing Histology with MulitPhoton microscopy (CHiMP, Applikate Technologies, Fairfield, CT), producing digital images of un- cut, un-embeded tissue to generate H&amp;E-like optical slices. Multiphoton images were downscaled to 0.5 um/px to match algorithm target resolution. CLAM models for subtyping RCC (chromophobe, clear cell, papillary) and LC (squamous &amp; adenocarcinoma) previously trained using TCGA and CPTAC whole slide images of physical slides were applied directly to CHiMP multiphoton images without adjustment. Reference cancer subtype classifications were provided from physical and digital slides. Results (if a Case Study enter NA) For the subtypes included during training, multiphoton WSIs of RCC and LC were accurately subtyped by the CLAM models without stain normalization nor network fine tuning producing high prediction levels. Subtypes not included during the training (namely oncocytoma for RCC) resulted in low scoring model predictions (below 0.85), indicating specificity of identification. Multiple slide levels improved interpretation of several difficult cases for CLAM predictions. Conclusion This preliminary data suggests that CLAM models trained on standard H&amp;E WSIs for RCC and LC subtyping are applicable to pseudo-H&amp;E multiphoton WSIs without domain adaptations. This implies that diagnostic histologic features have been learned by these CLAM models and are efficiently recognized in digital histology images produced via CHiMP.

3–4D soil model as challenge for future soil research: Quantitative soil modeling based on the solid phase

AbstractA 3–4D soil model represents a logical step forward from one‐dimensional soil columns (1D), two‐dimensional soil maps (2D), and three‐dimensional soil volumes (3D) toward dynamic soil models (4D), with time as the fourth dimension. The challenge is to develop modeling tools that account for the states of soil properties, including the spatial structure of solids and pores, as well as their dynamics, including soil mass and solute transfers in landscapes. Our envisioned 3–4D soil model approach aims at improving the capability to predict fundamental soil functions (e.g., plant growth, storage, matter fluxes) that provide ecosystem services in the socioeconomic context.This study provides a structured overview on current soil models, challenges, open questions, and urgent research needs for developing a 3–4D soil model. A 3–4D soil model should provide an inventory of spatially distributed and temporally variable soil properties. As basis for this, we propose a mass balance model for the solid phase, which needs to be supplemented by a model describing its structure. This should eventually provide adequate 3D parameter sets for the numerical modeling of soil functions (e.g., flow and transport). The target resolution is decameters in the horizontal plane and centimeters to decimeters in the vertical direction to represent characteristic soil properties and soil horizons. The actual state of soils and their properties can be estimated from spatial data that represent the soil forming factors, with the use of machine learning tools. Improved modeling of the dynamics of soil bulk density, biological processes, and the pore structure are required to relate the solid mass balance to matter fluxes.A 3–4D soil model can be built from several types of modeling approaches. We distinguish between (1) process models that simulate mass balances, fluxes and soil structure dynamics, (2) statistical pedometric models using machine learning and geostatistics to estimate the soil inventory within landscapes, and (3) pedotransfer functions to link observable attributes to specific model parameters required to simulate soil functions including water and matter fluxes. This should provide the prerequisites to predict the spatial distribution of soil functions and their changes in response to external forcing.This endeavor can draw upon many already established models and techniques, yet combining them into a newly created 3–4D soil model is a truly an ambitious, but promising task. The core of such a model is the bookkeeping of the solid mass together with soil structure, while accounting for biogeochemical and mechanical processes. The presented concepts are ambitious in context for research avenues toward the improvement of soil modeling by conjoining methods from a wide range of disciplines, including geological, geophysical, pedological, and remote sensing and visualization applications. The paper reviews and outlines research tools and needs for the 3‐D, spatially continuous representation of relevant soil properties and the modeling to represent the dynamics of soil properties and soil functions.

Spatial resolution analysis of passive bistatic radar using low-orbit navigation augmented signals

Aiming at the problem that the traditional monostatic ambiguity function is inconvenient to analyze the spatial resolution of the passive bistatic radar, this paper proposes a method to analyze it in a given airspace by using the spatial distribution of the ambiguity function. This method converts the time-delay Doppler distribution of the traditional ambiguity function into the spatial distribution near the reference target, which makes it more suitable for the intuitive analysis of the spatial resolution capability of the passive bistatic radar. This new method is used to analyze the target resolution capability in the designated airspace when Navigation augmented signals transmitted by the low-orbit communication satellite is used as the illuminator of opportunity. The distribution of the ambiguity function is transformed from a pushpin shape under the monostatic time delay-Doppler shift to a flat “sandwich biscuit” shape in space. The shape is closely related to the geometric relationship of the bistatic radar. Through the spatial distribution of the ambiguity function, the target resolution in a given airspace can be directly observed, which is convenient for analyzing the spatial resolution capability of the signal under different spatial geometric relationships. This also provides convenience for the optimal selection of the location of the illuminator of opportunity and receiving station.

Compact nano-void spectrometer based on a stable engineered scattering system

Random scattering of light in disordered media can be used for highly sensitive speckle-based wavemeters and spectrometers. However, the multiple scattering events that fold long optical paths within a compact space also make such devices exceedingly sensitive to vibrations and small disturbances to the disordered media. Here, we show how scattering can be engineered so that it can be used for a compact computational spectrometer that is largely insensitive to environmental factors. We designed and fabricated a three-dimensional pseudo-random nano-void pattern with 62% scattering efficiency. The controlled amount of multiple scattering ensured a sufficiently long optical path for the target resolution of 100 pm, with optimal long-term stability. The 200-μm-thick scattering silica substrate was integrated in a compact assembly with a low-cost camera sensor. The target resolution was achieved for full spectrum measurements while single wavelengths could be determined with 50 pm resolution. Such tailored scattering systems can improve the trade-off among cost, size, stability, and spectral resolution in computational spectrometers.

Photonic Sensor for Multiple Targets Detection under Adverse Weather Conditions in Autonomous Vehicles

Detection and tracing of multiple targets in a real-time scenario, particularly in the urban setup under adverse atmospheric conditions, has become a major challenge for autonomous vehicles (AVs). Photonic radars have emerged as promising candidates for Avs to realize via the recognition of traffic patterns, navigation, lane detection, self-parking, etc. In this work we developed a direct detection-based, frequency-modulated photonic radar to detect multiple stationary targets using four different transmission channels multiplexed over a single free space channel via wavelength division multiplexing (WDM). Additionally, the performance of the proposed photonic radar was examined under the impact of adverse weather conditions, such as rain and fog. The reported results in terms of received power and signal-to-noise ratio (SNR) showed successful detection of all the targets with bandwidths of 1 GHz and 4 GHz. The proposed system was also tested for range resolution of targets at 150 m and 6.75 cm resolution with 4 GHz bandwidth was reported, while resolution of 50 cm was reported with 1 GHz of bandwidth.

Temporally Stable Real-Time Joint Neural Denoising and Supersampling

Recent advances in ray tracing hardware bring real-time path tracing into reach, and ray traced soft shadows, glossy reflections, and diffuse global illumination are now common features in games. Nonetheless, ray budgets are still limited. This results in undersampling, which manifests as aliasing and noise. Prior work addresses these issues separately. While temporal supersampling methods based on neural networks have gained a wide use in modern games due to their better robustness, neural denoising remains challenging because of its higher computational cost. We introduce a novel neural network architecture for real-time rendering that combines supersampling and denoising, thus lowering the cost compared to two separate networks. This is achieved by sharing a single low-precision feature extractor with multiple higher-precision filter stages. To reduce cost further, our network takes low-resolution inputs and reconstructs a high-resolution denoised supersampled output. Our technique produces temporally stable high-fidelity results that significantly outperform state-of-the-art real-time statistical or analytical denoisers combined with TAA or neural upsampling to the target resolution.