Low-resolution Datasets Research Articles

Assessing Electronics with Advanced 3D X-ray Imaging Techniques, Nanoscale Tomography, and Deep Learning

This paper presents advanced workflows that combine 3D X-ray microscopy (XRM), nanoscale tomography, and deep learning (DL) to generate a detailed visualization of the interior of electronic devices and assemblies to enable the study of internal components for failure analysis (FA). Newly developed techniques, such as the integration of DL-based algorithms for 3D image reconstruction to improve scan quality through increased contrast and denoising, are also discussed in this article. In addition, a DL-based tool called DeepScout is presented. DeepScout uses 3D XRM scans in targeted regions of interest as training data for upscaling high-resolution in a low-resolution dataset, of a wider field of view, using a neural network model. Ultimately, these workflows can be run independently or complementary to other multiscale correlative microscopy evaluations, e.g., electron microscopy, and they will provide valuable insights into the inner workings of electronic packages and integrated circuits at multiple length scales, from macroscopic features on electronic devices (i.e., hundreds of mm) to microscopic details in electronic components (in the tens of nm). Understanding advanced electronic systems through X-ray imaging and machine learning—perhaps complemented with some additional correlative microscopy investigations—can speed development time, increase cost efficiency, and simplify FA and quality inspection of printed circuit boards (PCBs) and electronic devices assembled with new emerging technologies.

DGGI: Deep Generative Gradient Inversion with diffusion model

Federated learning is a privacy-preserving distributed framework that facilitates information fusion and sharing among different clients, enabling the training of a global model without exposing raw data. However, the gradient inversion attack that can reconstruct the training data via gradients has posed a significant threat. Prior attack approaches have demonstrated the efficacy of gradient inversion on low-resolution datasets with small batch sizes, which is impractical in real scenarios. To tackle this issue, this paper proposes an innovative and practical gradient inversion method, namely Deep Generative Gradient Inversion (DGGI), which employs the prior knowledge of diffusion models to enhance reconstruction performance on high-resolution datasets and larger batch sizes. Furthermore, a novel group consistency regularization term that constrains the distance between reconstruction and alignment images has been developed to address the issue of spatial variations caused by pre-trained diffusion model. Experiments conducted on both natural and medical image datasets demonstrate that our DGGI method outperforms state-of-the-art baselines in image reconstruction metrics. Furthermore, our approach achieves pixel-level reconstruction and causes leakage of privacy information, even at larger batch sizes or under various defenses, which can aid in the exploration of latent security concerns within information fusion models.

Anatomy-constrained synthesis for spleen segmentation improvement in unpaired mouse micro-CT scans with 3D CycleGAN

Objective. Auto-segmentation in mouse micro-CT enhances the efficiency and consistency of preclinical experiments but often struggles with low-native-contrast and morphologically complex organs, such as the spleen, resulting in poor segmentation performance. While CT contrast agents can improve organ conspicuity, their use complicates experimental protocols and reduces feasibility. We developed a 3D Cycle Generative Adversarial Network (CycleGAN) incorporating anatomy-constrained U-Net models to leverage contrast-enhanced CT (CECT) insights to improve unenhanced native CT (NACT) segmentation. Approach. We employed a standard CycleGAN with an anatomical loss function to synthesize virtual CECT images from unpaired NACT scans at two different resolutions. Prior to training, two U-Nets were trained to automatically segment six major organs in NACT and CECT datasets, respectively. These pretrained 3D U-Nets were integrated during the CycleGAN training, segmenting synthetic images, and comparing them against ground truth annotations. The compound loss within the CycleGAN maintained anatomical fidelity. Full image processing was achieved for low-resolution datasets, while high-resolution datasets employed a patch-based method due to GPU memory constraints. Automated segmentation was applied to original NACT and synthetic CECT scans to evaluate CycleGAN performance using the Dice Similarity Coefficient (DSC) and the 95th percentile Hausdorff Distance (HD95p). Main results. High-resolution scans showed improved auto-segmentation, with an average DSC increase from 0.728 to 0.773 and a reduced HD95p from 1.19 mm to 0.94 mm. Low-resolution scans benefited more from synthetic contrast, showing a DSC increase from 0.586 to 0.682 and an HD95p reduction from 3.46 mm to 1.24 mm. Significance. Implementing CycleGAN to synthesize CECT scans substantially improved the visibility of the mouse spleen, leading to more precise auto-segmentation. This approach shows the potential in preclinical imaging studies where contrast agent use is impractical.

Influence of high-resolution data on accurate curtailment loss estimation and optimal design of hybrid PV–wind power plants

Hybrid photovoltaic (PV) - wind power plants (HyPPs), i.e., where the PV and wind systems are co-located and share the Point of Interconnection (POI) with the grid, have recently attracted more attention. This trend is driven by the expected reduced capital and operational expenditures achieved through, e.g., shared land and POI infrastructure for HyPPs compared to two individual PV and wind installations. However, if the POI is underdimensioned relative to the PV and wind capacities, the generation from either or both of the assets must at times be curtailed, unless mitigated by solutions like energy storage. This curtailment might lead to income losses. Moreover, as HyPPs typically are designed using wind and PV generation data on hourly resolution, the actual curtailment losses can be underestimated. This might in turn lead to HyPP designs which are techno-economically sub-optimal.In this study, a comparative analysis is conducted to analyze how the curtailment and income loss estimations for HyPPs, as well as the techno-economic optimal HyPP topologies, are impacted by varying the input data time resolution. One year of site measured PV and wind power generation data on 5 s resolution from an existing HyPP located in Eastern Germany are used as basis for the study. For a HyPP topology with an undersized POI, where the installed capacities of the POI, PV, and wind systems are all equal, the curtailment losses are estimated to be 1.45%, 1.43% and 1.09% using the 5 s, 1 min and 1 h resolution datasets, respectively. Moreover, using the 1 h instead of the 1 min dataset leads to a 1.86% overestimation of the total Net Present Value (NPV) for this HyPP topology. As the shares of the PV and wind systems increase relative to the POI capacity, the differences in the estimation of the curtailment losses and NPVs between the high- and low-resolution datasets become more significant.

A novel method combining deep learning with the Kennard-Stone algorithm for training dataset selection for image-based rice seed variety identification.

Different varieties of rice vary in planting time, stress resistance, and other characteristics. With advances in rice-breeding technology, the number of rice varieties has increased significantly, making variety identification crucial for both trading and planting. This study collected RGB images of 20 hybrid rice seed varieties. An enhanced deep super-resolution network (EDSR) was employed to enhance image resolution, and a variety classification model utilizing the high-resolution dataset demonstrated superior performance to that of the model using the low-resolution dataset. A novel training sample selection methodology was introduced integrating deep learning with the Kennard-Stone (KS) algorithm. Convolutional neural networks (CNN) and autoencoders served as supervised and unsupervised feature extractors, respectively. The extracted feature vectors were subsequently processed by the KS algorithm to select training samples. The proposed methodologies exhibited superior performance over the random selection approach in rice variety classification, with an approximately 10.08% improvement in overall classification accuracy. Furthermore, the impact of noise on the proposed methodology was investigated by introducing noise to the images, and the proposed methodologies maintained superior performance relative to the random selection approach on the noisy image dataset. The experimental results indicate that both supervised and unsupervised learning models performed effectively as feature extractors, and the deep learning framework significantly influenced the selection of training set samples. This study presents a novel approach for training sample selection in classification tasks and suggests the potential for extending the proposed method to image datasets and other types of datasets. Further exploration of this potential is warranted. © 2024 Society of Chemical Industry.

Super-Resolution Image Reconstruction Method between Sentinel-2 and Gaofen-2 Based on Cascaded Generative Adversarial Networks

Due to the multi-scale and spectral features of remote sensing images compared to natural images, there are significant challenges in super-resolution reconstruction (SR) tasks. Networks trained on simulated data often exhibit poor reconstruction performance on real low-resolution (LR) images. Additionally, compared to natural images, remote sensing imagery involves fewer high-frequency components in network construction. To address the above issues, we introduce a new high–low-resolution dataset GF_Sen based on GaoFen-2 and Sentinel-2 images and propose a cascaded network CSWGAN combined with spatial–frequency features. Firstly, based on the proposed self-attention GAN (SGAN) and wavelet-based GAN (WGAN) in this study, the CSWGAN combines the strengths of both networks. It not only models long-range dependencies and better utilizes global feature information, but also extracts frequency content differences between different images, enhancing the learning of high-frequency information. Experiments have shown that the networks trained based on the GF_Sen can achieve better performance than those trained on simulated data. The reconstructed images from the CSWGAN demonstrate improvements in the PSNR and SSIM by 4.375 and 4.877, respectively, compared to the relatively optimal performance of the ESRGAN. The CSWGAN can reflect the reconstruction advantages of a high-frequency scene and provides a working foundation for fine-scale applications in remote sensing.

A Gold Standard-Derived Modular Barcoding Approach to Cancer Transcriptomics.

A challenge with studying cancer transcriptomes is in distilling the wealth of information down into manageable portions of information. In this resource, we develop an approach that creates and assembles cancer type-specific gene expression modules into flexible barcodes, allowing for adaptation to a wide variety of uses. Specifically, we propose that modules derived organically from high-quality gold standards such as The Cancer Genome Atlas (TCGA) can accurately capture and describe functionally related genes that are relevant to specific cancer types. We show that such modules can: (1) uncover novel gene relationships and nominate new functional memberships, (2) improve and speed up analysis of smaller or lower-resolution datasets, (3) re-create and expand known cancer subtyping schemes, (4) act as a "decoder" to bridge seemingly disparate established gene signatures, and (5) efficiently apply single-cell RNA sequencing information to other datasets. Moreover, such modules can be used in conjunction with native spreadsheet program commands to create a powerful and rapid approach to hypothesis generation and testing that is readily accessible to non-bioinformaticians. Finally, we provide tools for users to create and interpret their own modules. Overall, the flexible modular nature of the proposed barcoding provides a user-friendly approach to rapidly decoding transcriptome-wide data for research or, potentially, clinical uses.

RMSRGAN: A Real Multispectral Imagery Super-Resolution Reconstruction for Enhancing Ginkgo Biloba Yield Prediction

Multispectral remote sensing data with abundant spectral information can be used to compute vegetation indices to improve the accuracy of Ginkgo biloba yield prediction. The limited spatial resolution of multispectral cameras restricts the detail capture over wide farmland, but super-resolution (SR) reconstruction methods can enhance image quality. However, most existing SR models have been trained on images processed from downsampled high-resolution (HR) images, making them less effective in reconstructing real low-resolution (LR) images. This study proposes a GAN-based super-resolution reconstruction method (RMSRGAN) for multispectral remote sensing images of Ginkgo biloba trees in real scenes. A U-Net-based network is employed instead of the traditional discriminator. Convolutional block attention modules (CBAMs) are incorporated into the Residual-in-Residual Dense Blocks (RRDBs) of the generator and the U-Net of the discriminator to preserve image details and texture features. An unmanned aerial vehicle (UAV) equipped with a multispectral camera was employed to capture field multispectral remote sensing images of Ginkgo biloba trees at different spatial resolutions. Four matching HR and LR datasets were created from these images to train RMSRGAN. The proposed model outperforms the traditional models by achieving superior results in both quantitative evaluation metrics (peak signal-to-noise ratio (PSNR) is 32.490, 31.085, 27.084, 26.819, and structural similarity index (SSIM) is 0.894, 0.881, 0.832, 0.818, respectively) and qualitative evaluation visualization. Furthermore, the efficiency of our proposed method was tested by generating individual vegetation indices (VIs) from images taken before and after reconstruction to predict the yield of Ginkgo biloba. The results show that the SR images exhibit better R2 and RMSE values than LR images. These findings show that RMSRGAN can improve the spatial resolution of real multispectral images, increasing the accuracy of Ginkgo biloba yield prediction and providing more effective and accurate data support for crop management.

Giant sediment wave fields adjacent to debris-flow filled deep sea valleys: New evidence of cohesive flows transforming into dilute turbidity currents

Sediment waves are subaqueous sedimentary figures belonging to the supercritical flow domain and are of growing interest to the scientific community and industry. They are ubiquitously observed on the seafloor of world's oceans, as well as in the stratigraphic record imaged by marine seismic datasets. In this study we focus on the Cenozoic strata offshore Ivory Coast, where giant sediment waves developed at the base of slope range in height and wavelength: 10–100 m and 1–6 km, respectively. Sediment waves fields in this study developed simultaneously and adjacent to wide, rectilinear valleys, filled by mass-transport deposits. Thus, sediment waves serve as a rare example of large-scale deep-water cyclic steps formed through phase transformation (water entrainment and dilution) of laminar debris flows.The lithological nature of sediment waves can be estimated through the observation of polygonal faulting affecting the sediment waves fields, which suggest a dominant abundance of fine-grained material (clay and silt-prone). This study also shows that wide submarine valleys flanked by sediment waves do not necessarily correspond to sand-prone depositional systems, and that their potential to hold reservoir units for hydrocarbon exploration or CO2 storage should be evaluated with caution when in lower resolution datasets are used.

An Efficient Multi-Scale Attention Feature Fusion Network for 4K Video Frame Interpolation

Video frame interpolation aims to generate intermediate frames in a video to showcase finer details. However, most methods are only trained and tested on low-resolution datasets, lacking research on 4K video frame interpolation problems. This limitation makes it challenging to handle high-frame-rate video processing in real-world scenarios. In this paper, we propose a 4K video dataset at 120 fps, named UHD4K120FPS, which contains large motion. We also propose a novel framework for solving the 4K video frame interpolation task, based on a multi-scale pyramid network structure. We introduce self-attention to capture long-range dependencies and self-similarities in pixel space, which overcomes the limitations of convolutional operations. To reduce computational cost, we use a simple mapping-based approach to lighten self-attention, while still allowing for content-aware aggregation weights. Through extensive quantitative and qualitative experiments, we demonstrate the excellent performance achieved by our proposed model on the UHD4K120FPS dataset, as well as illustrate the effectiveness of our method for 4K video frame interpolation. In addition, we evaluate the robustness of the model on low-resolution benchmark datasets.

Low-dose GBCA administration for brain tumour dynamic contrast enhanced MRI: a feasibility study

A key limitation of current dynamic contrast enhanced (DCE) MRI techniques is the requirement for full-dose gadolinium-based contrast agent (GBCA) administration. The purpose of this feasibility study was to develop and assess a new low GBCA dose protocol for deriving high-spatial resolution kinetic parameters from brain DCE-MRI. Nineteen patients with intracranial skull base tumours were prospectively imaged at 1.5 T using a single-injection, fixed-volume low GBCA dose, dual temporal resolution interleaved DCE-MRI acquisition. The accuracy of kinetic parameters (ve, Ktrans, vp) derived using this new low GBCA dose technique was evaluated through both Monte-Carlo simulations (mean percent deviation, PD, of measured from true values) and an in vivo study incorporating comparison with a conventional full-dose GBCA protocol and correlation with histopathological data. The mean PD of data from the interleaved high-temporal-high-spatial resolution approach outperformed use of high-spatial, low temporal resolution datasets alone (p < 0.0001, t-test). Kinetic parameters derived using the low-dose interleaved protocol correlated significantly with parameters derived from a full-dose acquisition (p < 0.001) and demonstrated a significant association with tissue markers of microvessel density (p < 0.05). Our results suggest accurate high-spatial resolution kinetic parameter mapping is feasible with significantly reduced GBCA dose.

Downscaling Land Surface Temperature Derived from Microwave Observations with the Super-Resolution Reconstruction Method: A Case Study in the CONUS

Optical sensors cannot penetrate clouds and can cause serious missing data problems in optical-based Land Surface Temperature (LST) products. Under cloudy conditions, microwave observations are usually utilized to derive the land surface temperature. However, microwave sensors usually have coarse spatial resolutions. High-Resolution (HR) LST data products are usually desired for many applications. Instead of developing and launching new high-resolution satellite sensors for LST observations, a more economical and practical way is to develop proper methodologies to derive high-resolution LSTs from available Low-Resolution (LR) datasets. This study explores different algorithms to downscale low-resolution LST data to a high resolution. The existing regression-based downscaling methods usually require simultaneous observations and ancillary data. The Super-Resolution Reconstruction (SRR) method developed for traditional image enhancement can be applicable to high-resolution LST generation. For the first time, we adapted the SRR method for LST data. We specifically built a unique database of LSTs for the example-based SRR method. After deriving the LST data from the coarse-resolution passive microwave observations, the AMSR-E at 25 km and/or AMSR-2 at 10 km, we developed an algorithm to downscale them to a 1 km spatial resolution with the SRR method. The SRR downscaling algorithm can be implemented to obtain high-resolution LSTs without auxiliary data or any concurrent observations. The high-resolution LSTs are validated and evaluated with the ground measurements from the Surface Radiation (SURFRAD) Budget Network. The results demonstrate that the downscaled microwave LSTs have a high correlation coefficient of over 0.92, a small bias of less than 0.5 K, but a large Root Mean Square Error (RMSE) of about 4 K, which is similar to the original microwave LST, so the errors in the downscaled LST could have been inherited from the original microwave LSTs. The validation results also indicate that the example-based method shows a better performance than the self-similarity-based algorithm.

Sports Video Classification Method Based on Improved Deep Learning

Classifying sports videos is complex due to their dynamic nature. Traditional methods, like optical flow and the Histogram of Oriented Gradient (HOG), are limited by their need for expertise and lack of universality. Deep learning, particularly Convolutional Neural Networks (CNNs), offers more effective feature recognition in sports videos, but standard CNNs struggle with fast-paced or low-resolution sports videos. Our novel neural network model addresses these challenges. It begins by selecting important frames from sports footage and applying a fuzzy noise reduction algorithm to enhance video quality. The model then uses a bifurcated neural network to extract detailed features, leading to a densely connected neural network with a specific activation function for categorizing videos. We tested our model on a High-Definition Sports Video Dataset covering over 20 sports and a low-resolution dataset. Our model outperformed established classifiers like DenseNet, VggNet, Inception v3, and ResNet-50. It achieved high precision (0.9718), accuracy (0.9804), F-score (0.9761), and recall (0.9723) on the high-resolution dataset, and significantly better precision (0.8725) on the low-resolution dataset. Correspondingly, the highest values on the matrix of four traditional models are: precision (0.9690), accuracy (0.9781), F-score (0.9670), recall (0.9681) on the high-resolution dataset, and precision (0.8627) on the low-resolution dataset. This demonstrates our model’s superior performance in sports video classification under various conditions, including rapid motion and low resolution. It marks a significant step forward in sports data analytics and content categorization.

Degradation model and attention guided distillation approach for low resolution face recognition

Deep convolution neural networks (CNN) have shown their efficacy in face recognition tasks due to their ability to extract highly discriminant face representations from face images. On high-resolution benchmark datasets, outstanding identification and verification results have been achieved. However, the performance of these algorithms is significantly degraded when tested on low-resolution (LR) images such as those captured from surveillance cameras. A straightforward solution to this problem is to use both high-resolution (HR) images and corresponding down-sampled LR images during training. Although this strategy improves the performance for LR images, it has some limitations. First, there is a significant difference between down-sampled LR images and LR images from surveillance cameras, leading to performance saturation at an earlier stage. Another limitation is the deterioration in the performance of HR images. In this work, solutions to both these limitations are proposed. A degradation model is proposed that synthesizes LR images from corresponding HR, emulating the real-world degradation effects in synthetic data, thus enabling the face recognition system to tolerate various blurry and noisy effects. To address the deterioration in the performance of HR images, an attention-guided distillation is proposed, which utilizes attention maps from convolutional layers in combination with deep features to transfer informative HR features from teacher to student network. The attention maps from the teacher network guide the student network to a better optimum and produce resolution robust face representations. The results of the proposed approach on the popular LR datasets like SCface, Coxface, and PaSC show that it outperforms the recent state-of-the-art (SOTA) techniques by a significant margin demonstrating its effectiveness for different cross-resolution scenarios.

Kernel adaptive memory network for blind video super-resolution

Although recent video super-resolution (VSR) works show remarkable restoration performance for low-resolution (LR) video downscaled by a fixed known blur kernel, blind VSR suffers from severe performance degradation when the blur kernel is unknown. To alleviate this problem, blur kernel estimation methods have been proposed for VSR. However, existing VSR models must be trained separately with each LR dataset downscaled using all possible blur kernels. This is a time-consuming and memory-consuming task. To address these issues, we propose a kernel adaptive memory network for a blind VSR (KeMoVSR). The KeMoVSR mainly consists of a dual regression blur kernel estimator and a kernel adaptive VSR. The proposed blur kernel estimator predicts the parametric and non-parametric blur kernels by exploiting the blur kernel variation. Due to the blur kernel variation, the proposed kernel estimator can consider the temporal consistency of the blur kernel variation in adjacent frames, which leads to accurate blur kernel estimation in LR video frames. The proposed kernel adaptive VSR modulates the VSR weights according to the shape of the blur kernel by weight modulation layers. By integrating the proposed methods based on the memory network, we propose the KeMoVSR, which performs VSR by adaptively modulating the VSR weights using the blur kernel parameters as keys and values in memory networks. Experiments show that the KeMoVSR achieves superior performance compared to other blind VSR approaches. The KeMoVSR provides effective memory utilization that is appropriate for real-world scenarios. The code is available at https://github.com/dbseorms16/KeMoVSR.

Upsampling wave temporal resolution: Investigating wave parameters and the influence on WEC power performance

Power production of wave energy converters (WEC) predicted in the time domain utilize wave resource parameters and time-domain hydrodynamic model simulations of the WEC. While the hydrodynamic model provides high temporal resolution of power production (10’s of Hz), the wave resource parameters are often based on frequency-domain calculations with temporal resolution of 30 minutes to an hour. However, real ocean wave conditions vary on much shorter time scales. Relying on frequency-domain calculations will not be sufficient to capture the short-term variability and accurately predict WEC power production for a standardized methodology that follows power system requirements. These requirements need forecasted data with high sampling frequency for accurate energy predictions. Looking specifically at resource characterization, high temporal resolution datasets are not publicly available or do not exist for many coastal locations. Due to data availability, low temporal resolution datasets are being used in a majority of studies to generate representative wave conditions as inputs to numerical simulations. Representative wave conditions are used to generate wave spectrums. The issue with this practice is spectrums are then used to predict the efficiency of systems that will not accurately capture the variability of waves in short timeframes. Creating a standardized methodology to increase the temporal resolution of metaocean conditions to inform model development can provide better forecasting of power production. Random amplitude, Fourier coefficient methods have been suggested for WEC simulations of finite durations to improve the observed variability in wave heights and power production. Variability using this method does increase for finite durations compared to the commonly used deterministic amplitude method. In this paper we will investigate the influence of wave parameters (significant wave height, maximum wave height, and energy period) on the prediction of WEC power production. A better understanding of the influence of these parameters will provide a path towards future standardization methodology for resource inputs for time-domain modeling.

Physics-Informed Super-Resolution of Turbulent Channel Flows via Three-Dimensional Generative Adversarial Networks

For a few decades, machine learning has been extensively utilized for turbulence research. The goal of this work is to investigate the reconstruction of turbulence from minimal or lower-resolution datasets as inputs using reduced-order models. This work seeks to effectively reconstruct high-resolution 3D turbulent flow fields using unsupervised physics-informed deep learning. The first objective of this study is to reconstruct turbulent channel flow fields and verify these with respect to the statistics. The second objective is to compare the turbulent flow structures generated from a GAN with a DNS. The proposed deep learning algorithm effectively replicated the first- and second-order statistics of turbulent channel flows of Reτ= 180 within a 2% and 5% error, respectively. Additionally, by incorporating physics-based corrections to the loss functions, the proposed algorithm was also able to reconstruct λ2 structures. The results suggest that the proposed algorithm can be useful for reconstructing a range of 3D turbulent flows given computational and experimental efforts.

Mangrove Resource Mapping Using Remote Sensing in the Philippines: A Systematic Review and Meta-Analysis

In spite of their importance, mangroves are still threatened by a significant reduction in global forest cover due to conversion to non-forest land uses. To implement robust policies and actions in mangrove conservation, quantitative methods in monitoring mangrove attributes are vital. This study intends to study the trend in mangrove resource mapping using remote sensing (RS) to determine the appropriate methods and datasets to be used in monitoring the distribution, aboveground biomass (AGB), and carbon stock (CS) in mangroves. A meta-analysis of several research publications related to mangrove resource mapping using RS in the Philippines has been conducted. A database was constructed containing 59 peer-reviewed articles selected using the protocol search, appraisal, synthesis, analysis, report (PSALSAR) framework and preferred reporting items for systematic reviews and meta-analysis (PRISMA). The study clarified that support vector machine (SVM) has shown to be more effective (99%) in discriminating mangroves from other land cover. Light detection and ranging (LiDAR) data also has proven to give a promising result in overall accuracy in mangrove-extent mapping (99%), AGB, and CS estimates (99%), and even species-level mapping (77%). Medium to low-resolution datasets can still achieve high overall accuracy by using appropriate algorithms or predictive models such as the mangrove vegetation index (MVI). The study has also found out that there are still few reports on the usage of high-spatial-resolution datasets, most probably due to their commercial restrictions.

Correcting biases in tropical cyclone intensities in low-resolution datasets using dynamical systems metrics

Correcting biases in tropical cyclone intensities in low-resolution datasets using dynamical systems metrics

Demographic Bias in Low-Resolution Deep Face Recognition in the Wild

Face biometrics play a primary role in smart cities, from consumer- to organizational-level applications. This class of technologies has been recently shown to emphasize performance disparities across gender and ethnic groups. Prior work on demographic bias in deep face recognition has tended to focus exclusively on high-resolution images, leaving out low-resolution images captured, for example, by a certain surveillance camera at a distance. Notable reasons for this focus include the lack of low-resolution face image data sets that report user's demographic attributes. Therefore, demographic bias in low-resolution deep face recognition in the wild still remains under-explored. In this paper, we propose a framework for exploring demographic disparities in low-resolution face recognition at a distance. To this end, we devised a deep generative approach to turn high-resolution face images to their low-resolution counterpart. We then trained state-of-the-art face recognition models on different combinations of high- and low-resolution images. Finally, we assessed the model demographic disparities on artificially degraded images from five data sets. Our results show that face images artificially degraded through our generative approach are more realistic than those obtained with other statistical changes. Key disparities across gender and ethnic groups exist and urge timely interventions. Source code: <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://cutt.ly/FChLfOC</uri>