Dynamic Reconstruction Research Articles

High Dynamic Range Image Reconstruction from Saturated Images of Metallic Objects

This study considers a method for reconstructing a high dynamic range (HDR) original image from a single saturated low dynamic range (LDR) image of metallic objects. A deep neural network approach was adopted for the direct mapping of an 8-bit LDR image to HDR. An HDR image database was first constructed using a large number of various metallic objects with different shapes. Each captured HDR image was clipped to create a set of 8-bit LDR images. All pairs of HDR and LDR images were used to train and test the network. Subsequently, a convolutional neural network (CNN) was designed in the form of a deep U-Net-like architecture. The network consisted of an encoder, a decoder, and a skip connection to maintain high image resolution. The CNN algorithm was constructed using the learning functions in MATLAB. The entire network consisted of 32 layers and 85,900 learnable parameters. The performance of the proposed method was examined in experiments using a test image set. The proposed method was also compared with other methods and confirmed to be significantly superior in terms of reconstruction accuracy, histogram fitting, and psychological evaluation.

Deep learned triple-tracer multiplexed PET myocardial image separation

IntroductionIn multiplexed positron emission tomography (mPET) imaging, physiological and pathological information from different radiotracers can be observed simultaneously in a single dynamic PET scan. The separation of mPET signals within a single PET scan is challenging due to the fact that the PET scanner measures the sum of the PET signals of all the tracers. The conventional multi-tracer compartment modeling method (MTCM) requires staggered injections and assumes that the arterial input functions (AIFs) of each tracer are known.MethodsIn this work, we propose a deep learning-based method to separate triple-tracer PET images without explicitly knowing the AIFs. Dynamic triple-tracer noisy MLEM reconstruction was used as the network input and dynamic single-tracer noisy MLEM reconstructions were used as the training labels.ResultsA simulation study was performed to evaluate the performance of the proposed framework on triple-tracer ([18F]FDG+82Rb+[94mTc]sestamibi) PET myocardial imaging. The results show that the proposed methodology substantially reduced the noise level compared to the results obtained from single-tracer imaging. Additionally, it achieved lower bias and standard deviation in the separated single-tracer images compared to the MTCM-based method at both the voxel and ROI levels.DiscussionAs compared to the MTCM separation, the proposed method uses spatiotemporal information for separation, which enhances the separation performance at both the voxel and ROI levels. The simulation study also indicates the feasibility and potential of the proposed DL-based method for the application to pre-clinical and clinical studies.

Dynamic Unstructured Pruning Neural Network Image Super-resolution Reconstruction

Dynamic Unstructured Pruning Neural Network Image Super-resolution Reconstruction

Carbon Sequestration Dynamics in Peri-Urban Forests: Comparing Secondary Succession and Mature Stands under Varied Forest Management Practices

This study examines the impact of silvicultural and land-use management practices on carbon sequestration in peri-urban forest ecosystems, with a particular focus on human-induced carbon dynamics. The study area’s complex profile spans from a compact native forest to varying degrees of fragmentation. This included areas undergoing secondary succession forest without silvicultural interventions (No-SI) alongside sites subjected to high-intensity (High-SI) and low-intensity silvicultural interventions (Low-SI). The research assessed carbon stocks and sequestration in different carbon pools (living biomass, dead organic matter and soil) using field data, allometric equations and laboratory analysis. Findings reveal a significant correlation between the intensity of anthropogenic interventions and variations in carbon stocks. The CASMOFOR model facilitated the reconstruction of carbon stock and carbon-stock change dynamics over four decades (1980–2022), showing disparities in carbon storage capabilities linked to the structural characteristics of the sites. The Low-SI site had the highest carbon stock in all carbon pools (378 tonnes C ha−1), which is more than double compared to High-SI (161 tonnes C ha−1) or No-SI sites (134 tonnes C ha−1). However, the secondary succession forest (No-SI) demonstrated the highest annual carbon stock change (4.4 tonnes C ha−1 year−1), two times higher than the Low-SI mature stand (2.2 tonnes C ha−1 year−1), emphasising the resilience of forest ecosystems to recover and sustain carbon sequestration capacities after harvesting if forest land use remains unchanged. The study underscores the significant importance of anthropogenic interventions on carbon dynamics, especially for living tree biomass, which has consequences in enhancing carbon sequestration and contributing to emission reduction targets.

From Atomic-Level Synthesis to Device-Scale Reactors: A Multiscale Approach to Water Electrolysis.

ConspectusThe development of an advanced energy conversion system for water electrolysis with high efficiency and durability is of great significance for a hydrogen-powered society. This progress relies on the fabrication of electrocatalysts with superior electrochemical performance. Despite decades of advancements in exploring high-performance noble and non-noble metal electrocatalysts, several challenges persist at both the micro- and macrolevels in the field of water electrolysis.At the microlevel, which encompasses electrocatalyst synthesis and characterization, design strategies for high-performance electrocatalysts have primarily focused on interface chemical engineering. However, comprehensive understanding and investigation of interface chemical engineering across various length scales, from micrometers to atomic scales, are still lacking. This deficiency hampers the rational design of catalysts with optimal performance. Under harsh reaction conditions, such as high bias potential and highly acidic or alkaline media, the surface of catalyst materials is susceptible to undergoing "reconstruction", deviating from what is observed through ex situ characterization techniques postsynthesis. Conventional ex situ characterization methods do not provide an accurate depiction of the catalyst's structural evolution during the electrocatalytic reaction, hindering the exploration of the catalytic mechanism.At the macrolevel, pertaining to catalysis-performance evaluation systems and devices, traditional laboratory settings employ a conventional three-electrode or two-electrode system to assess the catalytic performance of electrocatalysts. However, this approach does not accurately simulate hydrogen production under realistic industrial conditions, such as elevated temperatures (60-70 °C), high current densities exceeding 0.5 A cm-2, and flowing electrolytes. To address this limitation, it is crucial to develop testing equipment and methodologies that replicate the actual industrial conditions.In this Account, we propose a multiscale research framework for water electrolysis, spanning from microscale synthesis to macroscale scaled reactor design. Our approach focuses on the design and evaluation of high-performance HER/OER (hydrogen evolution reaction/oxygen evolution reaction) electrocatalysts, incorporating the following strategies: Leveraging principles of interface chemical engineering across various length scales (micrometers, nanometers, and atoms) enables the design of catalyst materials that enhance both activity and durability. This approach provides a comprehensive understanding of the intricate interplay between the catalyst structure and activity, implementing in situ/operando characterization techniques to monitor dynamic interfacial reactions and surface reconstruction processes. This facilitates a profound exploration of catalytic reaction mechanisms, offering insights into the catalyst's structural evolution during the electrocatalytic reaction. We construct a laboratory-scale membrane electrode assembly (MEA) electrochemical reactor capable of operating at high current densities (>1 A cm-2) to evaluate the electrocatalytic performance under simulated industrial conditions. This ensures objective and authentic assessments of the catalyst application potential. Throughout the following sections, we illustrate the application of interface chemical engineering on different length scales in designing diverse electrocatalyst materials. We rely on in situ characterization techniques to gain a profound understanding of the mechanisms behind the HER and OER. Additionally, we describe the development of both acidic and alkaline MEA electrochemical reactors to enhance the precision of electrocatalytic performance evaluation. Finally, we provide a concise overview of the challenges and opportunities in this field.

Almost automorphic behaviours of nonlocal stochastic fuzzy Cohen–Grossberg lattice neural networks

Discrete-time Cohen–Grossberg neural networks (CGNNs) play important role in feedback neural networks. The existing literatures of CGNNs only regarded integral order discrete-time models without other variables. This paper builds the lattice model for nonlocal stochastic fuzzy CGNNs with reaction diffusions by employing the finite difference and Mittag–Leffler time Euler difference techniques. Further, the existence of a unique bounded almost automorphic sequence solution in distribution and global exponential convergence in the mean-square sense to the achieved difference model are investigated. At last, by using the tool of Matlab and time Euler difference for the Brownian motions, an illustrative example with simulations is used to show the feasible of the works of the current paper. We think that this work can contribute to achieve some effective real tasks, e.g. content-addressable memory, auto-association, dynamic reconstruction of a chaotic process, etc.

ADDRESSING THE INTENSITY OF CHANGES IN THE PREHISTORIC POPULATION DYNAMICS: POPULATION GROWTH RATE ESTIMATIONS IN THE CENTRAL BALKANS EARLY NEOLITHIC

ABSTRACT The intensity of changes in the population dynamics of the Early Neolithic (ca. 6250–5300 cal BC) communities in the Central Balkans was addressed by estimating the growth rate values. The Bayesian approach (Crema and Shoda 2021) of estimating intrinsic growth rates by fitting different models of population growth was applied to radiocarbon dates from the Early Neolithic sites in Serbia. We explored two possible episodes of population growth based on the results of the population dynamics reconstruction using the summed calibrated radiocarbon probability distributions (SPD) method. The results have shown that, within the first episode of growth, the intrinsic growth rate mean values are higher than the estimated continental average (between 1% and 2%). The results indicate a sudden and fast rise in population size, possibly due to the influx of the new population settling in the region at the beginning of the Neolithic. Lower values for the second episode could indicate more gradual population growth due to the mechanisms associated with the Neolithic Demographic Transition and the rise in fertility.

Accelerated Dynamic Reconstruction in Metal-Organic Frameworks with Ligand Defects for Selective Electrooxidation of Amines to Azos Coupling with Hydrogen Production.

Electrosynthesis coupled hydrogen production (ESHP) mostly involves catalyst reconstruction in aqueous phase, but accurately identifying and controlling the process is still a challenge. Herein, we modulated the electronic structure and exposed unsaturated sites of metal-organic frameworks (MOFs) via ligand defect to promote the reconstruction of catalyst for azo electrosynthesis (ESA) coupled with hydrogen production overall reaction. The monolayer Ni-MOFs achieved 89.8 % Faraday efficiency and 90.8 % selectivity for the electrooxidation of 1-methyl-1H-pyrazol-3-amine (Pyr-NH2) to azo, and an 18.5-fold increase in H2 production compared to overall water splitting. Operando X-ray absorption fine spectroscopy (XAFS) and various in situ spectroscopy confirm that the ligand defect promotes the potential dependent dynamic reconstruction of Ni(OH)2 and NiOOH, and the reabsorption of ligand significantly lowers the energy barrier of rate-determining step (*Pyr-NH to *Pyr-N). This work provides theoretical guidance for modulation of electrocatalyst reconstruction to achieve highly selective ESHP.

Accelerated Dynamic Reconstruction in Metal–Organic Frameworks with Ligand Defects for Selective Electrooxidation of Amines to Azos Coupling with Hydrogen Production

AbstractElectrosynthesis coupled hydrogen production (ESHP) mostly involves catalyst reconstruction in aqueous phase, but accurately identifying and controlling the process is still a challenge. Herein, we modulated the electronic structure and exposed unsaturated sites of metal–organic frameworks (MOFs) via ligand defect to promote the reconstruction of catalyst for azo electrosynthesis (ESA) coupled with hydrogen production overall reaction. The monolayer Ni‐MOFs achieved 89.8 % Faraday efficiency and 90.8 % selectivity for the electrooxidation of 1‐methyl‐1H‐pyrazol‐3‐amine (Pyr−NH2) to azo, and an 18.5‐fold increase in H2 production compared to overall water splitting. Operando X‐ray absorption fine spectroscopy (XAFS) and various in situ spectroscopy confirm that the ligand defect promotes the potential dependent dynamic reconstruction of Ni(OH)2 and NiOOH, and the reabsorption of ligand significantly lowers the energy barrier of rate‐determining step (*Pyr−NH to *Pyr−N). This work provides theoretical guidance for modulation of electrocatalyst reconstruction to achieve highly selective ESHP.

Multi-sensor data fusion reconstruction method for vibration dynamic responses of aerospace structures

The dynamic responses of key locations are important inputs for the life and reliability assessment of spacecraft structures. Due to the limited sensing resources, most critical responses are difficult to measure directly. A structural dynamic response reconstruction method is necessary. The responses of target locations can be reconstructed based on the empirical mode decomposition (EMD) of measured signals and the modal superposition. However, the structural modal information contained in the measured signal of a single sensor is limited, affecting the reconstruction accuracy. In this paper, a response reconstruction method based on multi-sensor data fusion is proposed. It is applied to a main load-bearing structure of a spacecraft and its typical components to verify its strain response reconstruction effect under random vibration loads. The experimental results show that multi-sensor data fusion improves the strain reconstruction accuracy. The maximum reduction in reconstruction error is from 8.7% to 1.3%. The reconstruction accuracy is further improved with the increase in the number of sensors. The optimal weighted fusion strategy for this problem is the weights defined by the Euclidean distance (EUC) or the dynamic time warping distance (DTW). The fusion results show a better performance with the increase in the power of the defined distance. The proposed multi-sensor fusion method improves the reconstruction accuracy via supplementing structural information to each other and eliminating the instability of single measured signals. More accurate dynamic responses via reconstruction reduce the large input uncertainty in life prediction and lay the foundation for building structural digital twins and managing structural health more effectively.

High dynamic range 3D shape reconstruction: Combination of adaptive fringe intensity, iterative exposure time adjustment and least quadratic curve fitting

For three-dimensional (3D) reconstruction of specular surfaces based on fringe projection profilometry (FPP), a hybrid approach combining adaptive fringe projection, iterative exposure time adjustment and least quadratic curve fitting is proposed. First, we propose an adaptive projection fringe intensity generation method in the camera image. Second, we build a new pixel matching between the camera and projector, and correct the pixel matching on the specular surfaces based on linear interpolation. Third, we propose an iterative exposure time adjustment method based on saturated pixel statistics. Fourth, we fuse the multiple results to obtain the final 3D reconstructed surface. Finally, we adopt a least quadratic curve fitting algorithm to fit the residual phase errors. Experiments confirmed the validity of the proposed method.

Using multiobjective optimization to reconstruct interferometric data

Context. In very long baseline interferometry (VLBI), signals recorded at multiple antennas are combined to form a sparsely sampled virtual aperture with an effective diameter set by the largest separation between the antennas. Due to the sparsity of the sampled aperture, VLBI imaging constitutes an ill-posed inverse problem. Various algorithms have been employed to deal with the VLBI imaging, including the recently proposed multiobjective evolutionary algorithm by decomposition (MOEA/D) described in the first paper of this series. Aims. Among the approaches to the reconstruction of the image features in total intensity from sparsely sampled visibilities, extensions to the polarimetric and the temporal domain are of great interest for the VLBI community in general and the Event Horizon Telescope Collabroration (EHTC) in particular. Based on the success of MOEA/D in presenting an alternative claim of the image structure in a unique, fast, and largely unsupervised way, we study the extension of MOEA/D to polarimetric and time dynamic reconstructions in this paper. Methods. To this end, we utilized the multiobjective, evolutionary framework introduced for MOEA/D, but added the various penalty terms specific to total intensity imaging time-variable and polarimetric variants, respectively. We computed the Pareto front (the sample of all non-dominated solutions) and identified clusters of close proximities. Results. We tested MOEA/D with synthetic data sets that are representative for the main science targets and instrumental configuration of the EHTC and its possible successors. We successfully recovered the polarimetric and time-dynamic signature of the ground truth movie (even with relative sparsity) and a set of realistic data corruptions. Conclusions. MOEA/D has been successfully extended to polarimetric and time-dynamic reconstructions and, specifically, in a setting that would be expected for the EHTC. It offers a unique alternative and independent claim to the already existing methods, along with a number of additional benefits, namely: it is the first method that effectively explores the problem globally and compared to regularized maximum likelihood (RML) methods. Thus, it waives the need for parameter surveys. Hence, MOEA/D is a novel, useful tool to characterize the polarimetric and dynamic signatures in a VLBI data set robustly with a minimal set of user-based choices. In a consecutive work, we will address the last remaining limitation for MOEA/D (the number of pixels and numerical performance), so that MOEA/D can firmly solidify its place within the VLBI data reduction pipeline.

Hyperspectral image dynamic range reconstruction using deep neural network-based denoising methods

Hyperspectral (HS) measurement is among the most useful tools in agriculture for early disease detection. However, the cost of HS cameras that can perform the desired detection tasks is prohibitive-typically fifty thousand to hundreds of thousands of dollars. In a previous study at the Agricultural Research Organization’s Volcani Institute (Israel), a low-cost, high-performing HS system was developed which included a point spectrometer and optical components. Its main disadvantage was long shooting time for each image. Shooting time strongly depends on the predetermined integration time of the point spectrometer. While essential for performing monitoring tasks in a reasonable time, shortening integration time from a typical value in the range of 200 ms to the 10 ms range results in deterioration of the dynamic range of the captured scene. In this work, we suggest correcting this by learning the transformation from data measured with short integration time to that measured with long integration time. Reduction of the dynamic range and consequent low SNR were successfully overcome using three developed deep neural networks models based on a denoising auto-encoder, DnCNN and LambdaNetworks architectures as a backbone. The best model was based on DnCNN using a combined loss function of ℓ2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\ell _{2}$$\\end{document} and Kullback–Leibler divergence on images with 20 consecutive channels. The full spectrum of the model achieved a mean PSNR of 30.61 and mean SSIM of 0.9, showing total improvement relatively to the 10 ms measurements’ mean PSNR and mean SSIM values by 60.43% and 94.51%, respectively.

Nonlocal based FISTA network for noninvasive cardiac transmembrane potential imaging

Objective. The primary aim of our study is to advance our understanding and diagnosis of cardiac diseases. We focus on the reconstruction of myocardial transmembrane potential (TMP) from body surface potential mapping. Approach. We introduce a novel methodology for the reconstruction of the dynamic distribution of TMP. This is achieved through the integration of convolutional neural networks with conventional optimization algorithms. Specifically, we utilize the subject-specific transfer matrix to describe the dynamic changes in TMP distribution and ECG observations at the body surface. To estimate the TMP distribution, we employ LNFISTA-Net, a learnable non-local regularized iterative shrinkage-thresholding network. The coupled estimation processes are iteratively repeated until convergence. Main results. Our experiments demonstrate the capabilities and benefits of this strategy. The results highlight the effectiveness of our approach in accurately estimating the TMP distribution, thereby providing a reliable method for the diagnosis of cardiac diseases. Significance. Our approach demonstrates promising results, highlighting its potential utility for a range of applications in the medical field. By providing a more accurate and dynamic reconstruction of TMP, our methodology could significantly improve the diagnosis and treatment of cardiac diseases, thereby contributing to advancements in healthcare.

Lepidopteran scales in lake sediments as a reliable proxy for spruce budworm outbreak events in the boreal forest of Eastern Canada

Characterizing disturbance regimes over long time scales is paramount for describing and identifying their variability. The most important biotic disturbance in the eastern Canadian boreal forest is the defoliation caused by the eastern spruce budworm, a moth of the insect order Lepidoptera. Lepidopteran scales have recently been used to reconstruct spruce budworm population fluctuations throughout the Holocene. However, this novel proxy has yet to be compared to an independent proxy. This study aimed to determine whether lepidopteran scales found in the surface sediments of boreal lakes tracked large spruce budworm populations, that is, outbreaks, using yearly aerial surveys (1967–present) of spruce budworm defoliation as an independent proxy. Scales were extracted (1 cm resolution) from the top 20 cm of 210Pb-dated sediment cores recovered from nine lakes. To identify significant abundance peaks of scales in the time series, we removed background noise using a modified version of CharAnalysis. A 100-year smoothing window width combined with a 60th percentile threshold yielded the highest true positive and true negative occurrences, and the lowest false positive and false negative occurrences, with values of 0.69 and 0.70 for Cohen’s Kappa and Matthews correlation coefficient, respectively. Our findings demonstrate that lepidopteran scales are a suitable proxy for identifying spruce budworm outbreaks in the sediment record enabling the reconstruction of budworm and other lepidopteran species outbreak dynamics at millennial timescales.

Three Dimensional Reconstruction of Two-step Moving Objects based on Phase-shifting Profilometry

In recent years, 3D object reconstruction based on phase-shifting profilometry has gradually received attention and been widely applied. Domestic and foreign scholars have been continuously researching and exploring the accuracy and speed of three-dimensional measurement, and gradually developing towards dynamic measurement. Most dynamic measurements require projecting multiple stripe patterns to obtain sufficient object phase information, and the more stripes there are, the greater the phase error caused by motion. This article proposes the use of increasing the sampling fringe pattern during the projection period of a fringe pattern to achieve high frame rate dynamic 3D object reconstruction. By combining the intensity values of ambient light; Finally, the phase information of the object is extracted by tracking the motion information obtained from the moving object. This article demonstrates the feasibility of this method through simulation experiments and improves the frame rate of 3D reconstruction of moving objects.

Low‐coordination Nanocrystalline Copper‐based Catalysts through Theory‐guided Electrochemical Restructuring for Selective CO2 Reduction to Ethylene

AbstractRevealing the dynamic reconstruction process and tailoring advanced copper (Cu) catalysts is of paramount significance for promoting the conversion of CO2 into ethylene (C2H4), paving the way for carbon neutralization and facilitating renewable energy storage. In this study, we initially employed density functional theory (DFT) and molecular dynamics (MD) simulations to elucidate the restructuring behavior of a catalyst under electrochemical conditions and delineated its restructuring patterns. Leveraging insights into this restructuring behavior, we devised an efficient, low‐coordination copper‐based catalyst. The resulting synthesized catalyst demonstrated an impressive Faradaic efficiency (FE) exceeding 70 % for ethylene generation at a current density of 800 mA cm−2. Furthermore, it showed robust stability, maintaining consistent performance for 230 hours at a cell voltage of 3.5 V in a full‐cell system. Our research not only deepens the understanding of the active sites involved in designing efficient carbon dioxide reduction reaction (CO2RR) catalysts but also advances CO2 electrolysis technologies for industrial application.

Low-coordination Nanocrystalline Copper-based Catalysts through Theory-guided Electrochemical Restructuring for Selective CO2 Reduction to Ethylene.

Revealing the dynamic reconstruction process and tailoring advanced copper (Cu) catalysts is of paramount significance for promoting the conversion of CO2 into ethylene (C2H4), paving the way for carbon neutralization and facilitating renewable energy storage. In this study, we initially employed density functional theory (DFT) and molecular dynamics (MD) simulations to elucidate the restructuring behavior of a catalyst under electrochemical conditions and delineated its restructuring patterns. Leveraging insights into this restructuring behavior, we devised an efficient, low-coordination copper-based catalyst. The resulting synthesized catalyst demonstrated an impressive Faradaic efficiency (FE) exceeding 70 % for ethylene generation at a current density of 800 mA cm-2. Furthermore, it showed robust stability, maintaining consistent performance for 230 hours at a cell voltage of 3.5 V in a full-cell system. Our research not only deepens the understanding of the active sites involved in designing efficient carbon dioxide reduction reaction (CO2RR) catalysts but also advances CO2 electrolysis technologies for industrial application.

Optimizing the intermediates adsorbability and revealing the dynamic reconstruction of Co6Fe3S8 solid solution for bifunctional water splitting

Co9S8 has been extensively studied as a promising catalyst for water electrolysis. Doping Co9S8 with Fe improves its oxygen evolution reaction (OER) performance by regulating the catalyst self-reconfigurability and enhancing the absorption capacity of OER intermediates. However, the poor alkaline hydrogen evolution reaction (HER) properties of Co9S8 limit its application in bifunctional water splitting. Herein, we combined Fe doping and sulfur vacancy engineering to synergistically enhance the bifunctional water-splitting performance of Co9S8. The as-synthesized Co6Fe3S8 catalyst exhibited excellent OER and HER characteristics with low overpotentials of 250 and 84 mV, respectively. It also resulted in the low Tafel slopes of 135 mV dec−1 for the OER and 114 mV dec−1 for the HER. A two-electrode electrolytic cell with Co6Fe3S8 used as both the cathode and anode produced a current density of 10 mA cm−2 at a low voltage of only 1.48 V, maintaining high stability for 100 h. The results of in/ex-situ experiments indicated that the OER process induced electrochemical reconfiguration, forming CoOOH/FeOOH active species on the catalyst surface to enhance its OER performance. Density functional theory (DFT) simulations revealed that Fe doping and the presence of unsaturated coordination metal sites in Co6Fe3S8 promoted H2O and H* adsorption for the HER. The findings of this study can help develop a strategy for designing highly efficient bifunctional water splitting electrocatalysts.

Dynamic Reconstruction of Two-Dimensional Defective Bi Nanosheets for Efficient Electrocatalytic Urea Synthesis.

Catalyst surface dynamics drive the generation of active species for electrocatalytic reactions. Yet, the understanding of dominant site formation and reaction mechanisms is limited. In this study, we thoroughly investigate the dynamic reconstruction of two-dimensional defective Bi nanosheets from exfoliated Bi2Se3 nanosheets under electrochemical CO2 and nitrate (NO3 -) reduction conditions. The ultrathin Bi2Se3 nanosheets obtained by NaBH4-assisted cryo-mediated liquid-phase exfoliation are more easily reduced and reconstructed to Bi nanosheets with high-density grain boundaries (GBs; GB-rich Bi). The reconstructed GB-rich Bi catalyst affords a remarkable yield rate of 4.6 mmol h-1 mgcat. -1 and Faradaic efficiency of 32 % for urea production at -0.40 V vs. RHE. Notably, this yield rate is 2 and 8.2 times higher than those of the low-GB Bi and bulk Bi catalysts, respectively. Theoretical analysis demonstrates that the GB sites significantly reduce the *CO and *NH2 intermediate formation energy and C-N coupling energy barrier, enabling selective urea electrosynthesis on the GB-rich Bi catalyst. This work will trigger further research into the structure-activity interplay in dynamic processes using in situ techniques.