Reliable image dehazing by NeRF.

Image dehazing is a typical low-level visual task. With the continuous improvement of network performance and the introduction of various prior knowledge, the ability of image dehazing is becoming stronger. However, the existing dehazing methods have problems such as the inability to obtain real shooting datasets, unreliable dehazing processes, and the difficulty to deal with complex lighting scenes. To solve these problems, we propose a new haze model combining the optical scattering model and the computer graphics rendering. Based on the new haze model, we propose a high-quality and widely applicable dehazing dataset generation pipeline that does not require paired-data training and prior knowledge. We reconstruct the three-dimensional fog space with array camera and remove haze by thresholding voxel deletion. We use the Unreal Engine 5 to generate simulation datasets and the real shooting in laboratory to verify the effectiveness and the reliability of our generation pipeline. Through our pipeline, we can obtain wonderful dehaze results and dehaze datasets under various complex outdoors lighting conditions. We also propose a dehaze dataset enhancement method based on voxel control. Our pipeline and data enhancement are suitable for the latest algorithm model, these solutions can obtain better visual effects and objective indicators.

Wavelength encoding spectral imaging based on the combination of deeply learned filters and an RGB camera.

Hyperspectral imaging is a critical tool for gathering spatial-spectral information in various scientific research fields. As a result of improvements in spectral reconstruction algorithms, significant progress has been made in reconstructing hyperspectral images from commonly acquired RGB images. However, due to the limited input, reconstructing spectral information from RGB images is ill-posed. Furthermore, conventional camera color filter arrays (CFA) are designed for human perception and are not optimal for spectral reconstruction. To increase the diversity of wavelength encoding, we propose to place broadband encoding filters in front of the RGB camera. In this condition, the spectral sensitivity of the imaging system is determined by the filters and the camera itself. To achieve an optimal encoding scheme, we use an end-to-end optimization framework to automatically design the filters' transmittance functions and optimize the weights of the spectral reconstruction network. Simulation experiments show that our proposed spectral reconstruction network has excellent spectral mapping capabilities. Additionally, our novel joint wavelength encoding imaging framework is superior to traditional RGB imaging systems. We develop the deeply learned filter and conduct actual shooting experiments. The spectral reconstruction results have an attractive spatial resolution and spectral accuracy.

Enhancing the action of serotonin by three different mechanisms prevents spontaneous seizure-induced mortality in Dravet mice.

OBJECTIVE: Sudden unexpected death in epilepsy (SUDEP) is an underestimated complication of epilepsy. Previous studies have demonstrated that enhancement of serotonergic neurotransmission suppresses seizure-induced sudden death in evoked seizure models. However, it is unclear whether elevated serotonin (5-HT) function will prevent spontaneous seizure-induced mortality (SSIM), which is characteristic of human SUDEP. We examined the effects of 5-HT-enhancing agents that act by three different pharmacological mechanisms on SSIM in Dravet mice, which exhibit a high incidence of SUDEP, modeling human Dravet syndrome. METHODS: Dravet mice of both sexes were evaluated for spontaneous seizure characterization and changes in SSIM incidence induced by agents that enhance 5-HT-mediated neurotransmission. Fluoxetine (a selective 5-HT reuptake inhibitor), fenfluramine (a 5-HT releaser and agonist), SR 57227 (a specific 5-HT3 receptor agonist), or saline (vehicle) was intraperitoneally administered over an 8-day period in Dravet mice, and the effect of these treatments on SSIM was examined. RESULTS: Spontaneous seizures in Dravet mice generally progressed from wild running to tonic seizures with or without SSIM. Fluoxetine at 30 mg/kg, but not at 20 or 5 mg/kg, significantly reduced SSIM compared with the vehicle control. Fenfluramine at 1-10 mg/kg, but not .2 mg/kg, fully protected Dravet mice from SSIM, with all mice surviving. Compared with the vehicle control, SR 57227 at 20 mg/kg, but not at 10 or 5 mg/kg, significantly lowered SSIM. The effect of these drugs on SSIM was independent of sex. SIGNIFICANCE: Our data demonstrate that elevating serotonergic function by fluoxetine, fenfluramine, or SR 57227 significantly reduces or eliminates SSIM in Dravet mice in a sex-independent manner. These findings suggest that deficits in serotonergic neurotransmission likely play an important role in the pathogenesis of SSIM, and fluoxetine and fenfluramine, which are US Food and Drug Administration-approved medications, may potentially prevent SUDEP in at-risk patients.

Adhesion between EVs and tumor cells facilitated EV-encapsulated doxorubicin delivery via ICAM1.

Doxorubicin (Dox) is an anti-tumor drug with a broad spectrum, whereas the cardiotoxicity limits its further application. In clinical settings, liposome delivery vehicles are used to reduce Dox cardiotoxicity. Here, we substitute extracellular vesicles (EVs) for liposomes and deeply investigate the mechanism for EV-encapsulated Dox delivery. The results demonstrate that EVs dramatically increase import efficiency and anti-tumor effects of Dox in vitro and in vivo, and the efficiency increase benefits from its unique entry pattern. Dox-loading EVs repeat a "kiss-and-run" motion before EVs internalization. Once EVs touch the cell membrane, Dox disassociates from EVs and directly enters the cytoplasm, leading to higher and faster Dox import than single Dox. This unique entry pattern makes the adhesion between EVs and cell membrane rather than the total amount of EV internalization the key factor for regulating the Dox import. Furthermore, we recognize ICAM1 as the molecule mediating the adhesion between EVs and cell membranes. Interestingly, EV-encapsulated Dox can induce ICAM1 expression by irritating IFN-γ and TNF-α secretion in TME, thereby increasing tumor targeting of Dox-loading EVs. Altogether, EVs and EV-encapsulated Dox synergize via ICAM1, which collectively enhances the curative effects for tumor treatment.

Efficient electrochemical oxidation of antibiotic wastewater using a graphene-loaded PbO2 membrane anode: Mechanisms and applications.

This paper aims to develop a flow-through electrochemical system with a series of graphene nanoparticles loaded PbO2 reactive electrochemical membrane electrodes (GNPs-PbO2 REMs) on porous Ti substrates with pore sizes of 100, 150, 300 and 600 µm, and apply them to treat antibiotic wastewater. Among them, the GNPs-PbO2 with Ti substrate of 150 µm (Ti-150/GNPs-PbO2) had superior electrochemical degradation performance over the REMs with other pore sizes due to its smaller crystal size, larger electrochemical active specific area, lower charge-transfer impedance and larger oxygen evolution potential. Under the relatively optimized conditions of initial pH of 5, current density of 15 mA cm-2, and membrane flux of 4.20 m3 (m2·h)-1, the Ti-150/GNPs-PbO2 REM realized 99.34% of benzylpenicillin sodium (PNG) removal with an EE/O of 6.52 kWh m-3. Its excellent performance could be explained as the increased mass transfer. Then three plausible PNG degradation pathways in the flow-through electrochemical system were proposed, and great stability and safety of Ti-150/GNPs-PbO2 REM were demonstrated. Moreover, a single-pass Ti-150/GNPs-PbO2 REM system with five-modules in series was designed, which could consistently treat real antibiotic wastewater in compliance with disposal requirements of China. Thus, this study evidenced that the flow-through electrochemical system with the Ti-150/GNPs-PbO2 REM is an efficient alternative for treating antibiotic wastewater.

Enhancing biocathode denitrification performance with nano-Fe3O4 under polarity period reversal.

The presence of excessive concentrations of nitrate poses a threat to both the environment and human health, and the bioelectrochemical systems (BESs) are attractive green technologies for nitrate removal. However, the denitrification efficiency in the BESs is still limited by slow biofilm formation and nitrate removal. In this work, we demonstrate the efficacy of novel combination of magnetite nanoparticles (nano-Fe3O4) with the anode-cathode polarity period reversal (PPR-Fe3O4) for improving the performance of BESs. After only two-week cultivation, the highest cathodic current density (7.71 ± 1.01 A m-2) and NO3--N removal rate (8.19 ± 0.97 g m-2 d-1) reported to date were obtained in the PPR-Fe3O4 process (i.e., polarity period reversal with nano-Fe3O4 added) at applied working voltage of -0.2 and -0.5 V (vs Ag/AgCl) under bioanodic and biocathodic conditions, respectively. Compared with the polarity reversal once only process, the PPR process (i.e., polarity period reversal in the absence of nano-Fe3O4) enhanced bioelectroactivity through increasing biofilm biomass and altering microbial community structure. Nano-Fe3O4 could enhance extracellular electron transfer as a result of promoting the formation of extracellular polymers containing Fe3O4 and reducing charge transfer resistance of bioelectrodes. This work develops a novel biocathode denitrification strategy to achieve efficient nitrate removal after rapid cultivation.

Research progress of inland river water quality monitoring technology based on unmanned aerial vehicle hyperspectral imaging technology.

In recent years, increasing demand for inland river water quality precision management has heightened the necessity for real-time, rapid, and continuous monitoring of water conditions. By analyzing the optical properties of water bodies remotely, unmanned aerial vehicle (UAV) hyperspectral imaging technology can assess water quality without direct contact, presenting a novel method for monitoring river conditions. However, there are currently some challenges to this technology that limit the promotion application of this technology, such as underdeveloped sensor calibration, atmospheric correction algorithms, and limitations in modeling non-water color parameters. This article evaluates the advantages and disadvantages of traditional sensor calibration methods and considers factors like sensor aging and adverse weather conditions that impact calibration accuracy. It suggests that future improvements should target hardware enhancements, refining models, and mitigating external interferences to ensure precise spectral data acquisition. Furthermore, the article summarizes the limitations of various traditional atmospheric correction methods, such as complex computational requirements and the need for multiple atmospheric parameters. It discusses the evolving trends in this technology and proposes streamlining atmospheric correction processes by simplifying input parameters and establishing adaptable correction algorithms. Simplifying these processes could significantly enhance the accuracy and feasibility of atmospheric correction. To address issues with the transferability of water quality inversion models regarding non-water color parameters and varying hydrological conditions, the article recommends exploring the physical relationships between spectral irradiance, solar zenith angle, and interactions with water constituents. By understanding these relationships, more accurate and transferable inversion models can be developed, improving the overall effectiveness of water quality assessment. By leveraging the sensitivity and versatility of hyperspectral sensors and integrating interdisciplinary approaches, a comprehensive database for water quality assessment can be established. This database enables rapid, real-time monitoring of non-water color parameters which offers valuable insights for the precision management of inland river water quality.

Effect of UV pretreatment on the source control of floR during subsequent biotreatment of florfenicol wastewater.

UV photolysis has been recommended as an alternative pretreatment method for the elimination of antibacterial activity of antibiotics against the indicator strain, but the pretreated antibiotic intermediates might not lose their potential to induce antibiotic resistance genes (ARGs) proliferation during subsequent biotreatment processes. The presence of florfenicol (FLO) in wastewater seriously inhibits the metabolic performance of anaerobic sludge microorganisms, especially the positive correlation between UV irradiation doses and ATP content, while it did not significantly affect the organics utilization ability and protein biosynthetic process of aerobic microorganisms. After sufficient UV pretreatment, the relative abundances of floR from genomic or plasmid DNA in subsequent aerobic and anaerobic biotreatment processes both decreased by two orders of magnitude, maintained at the level of the groups without FLO selective pressure. Meanwhile, the abundances of floR under anaerobic condition were always lower than that under aerobic condition, suggesting that anaerobic biotreatment systems might be more suitable for the effective control of target ARGs. The higher abundance of floR in plasmid DNA than in genome also indicated that the potential transmission risk of mobile ARGs should not be ignored. In addition, the relative abundance of intI1 was positively correlated with floR in its corresponding genomic or plasmid DNA (p < 0.05), which also increased the potential horizontal transfer risk of target ARGs. This study provides new insights into the effect of preferential UV photolysis as a pretreatment method for the enhancement of metabolic performance and source control of target ARGs in subsequent biotreatment processes. KEY POINTS: • Sufficient UV photolytic pretreatment efficiently controlled the abundance of floR • A synchronous decrease in abundance of intI1 reduced the risk of horizontal transfer • An appreciable abundance of floR in plasmid DNA was a potential source of total ARGs.

Analysis of swallowing and voice-related quality of life in patients after supracricoid partial laryngectomy.

OBJECTIVE: This study evaluated the swallowing and voice function of laryngeal cancer patients after Supracricoid Partial Laryngectomy(SCPL), and its influence on quality of life to provide a reference for the selection of surgical methods for laryngeal cancer patients. METHODS: Twenty-one patients who received SCPL between April 2015 and November 2021 were included. Each patient's swallowing function and quality of life were assessed through fiberoptic endoscopic examination of swallowing (FEES) and the M.D. Anderson Dysphagia Inventory (MDADI). Fundamental, jitter, shimmer, maximum phonation time (MPT), and voice handicap index-10 (VHI-10) were performed to assess voice function and voice-related quality of life. RESULTS: The results of the FEES of the 21 patients were as follows: the rates of pharyngeal residue after swallowing solid, semiliquid, and liquid food were 0%, 28.57%, and 38.09%, respectively; the rates of laryngeal infiltration after swallowing solid, semiliquid, and liquid food were 0%, 28.57%, and 4.76%, respectively; and aspiration did not occur in any of the patients. In the evaluation of swallowing quality of life, the mean total MDADI score was 92.6 ± 6.32. The voice function evaluation showed that the mean F0, jitter, shimmer, and MPT values were 156.01 ± 120.87 (HZ), 11.57 ± 6.21 (%), 35.37 ± 14.16 (%) and 7.85 ± 6.08 (s), respectively. The mean total VHI-10 score was 7.14 ± 4.84. CONCLUSION: SCPL provides patients with satisfactory swallowing and voice function. The patients in this study were satisfied with their quality of life in terms of swallowing and voice. SCPL can be used as a surgical method to preserve laryngeal function in patients with laryngeal cancer.

Elimination of varying chromatic aberrations based on diffractive optics.

In telescopic systems consisting of Alvarez lenses, chromatic aberrations vary with the magnifications and the fields of view. Computational imaging has developed rapidly in recent years, therefore we propose a method of optimizing the DOE and the post-processing neural network in 2 stages for achromatic aberrations. We apply the iterative algorithm and the gradient descent method to optimize the DOE, respectively, and then adopt U-Net to further optimize the results. The results show that the optimized DOEs improve the results, the gradient descent optimized DOE with U-Net performs the best and has a very robust and good performance in the case of simulated chromatic aberrations. The results also verify the validity of our algorithm.

Snapshot hyperspectral imaging based on equalization designed DOE.

Hyperspectral imaging attempts to determine distinctive information in spatial and spectral domain of a target. Over the past few years, hyperspectral imaging systems have developed towards lighter and faster. In phase-coded hyperspectral imaging systems, a better coding aperture design can improve the spectral accuracy relatively. Using wave optics, we post an equalization designed phase-coded aperture to achieve desired equalization point spread functions (PSFs) which provides richer features for subsequent image reconstruction. During the reconstruction of images, our raised hyperspectral reconstruction network, CAFormer, achieves better results than the state-of-the-art networks with less computation by substituting self-attention with channel-attention. Our work revolves around the equalization design of the phase-coded aperture and optimizes the imaging process from three aspects: hardware design, reconstruction algorithm, and PSF calibration. Our work is putting snapshot compact hyperspectral technology closer to a practical application.

Image restoration for optical zooming system based on Alvarez lenses.

Alvarez lenses are known for their ability to achieve a broad range of optical power adjustment by utilizing complementary freeform surfaces. However, these lenses suffer from optical aberrations, which restrict their potential applications. To address this issue, we propose a field of view (FOV) attention image restoration model for continuous zooming. In order to simulate the degradation of optical zooming systems based on Alvarez lenses (OZA), a baseline OZA is designed where the polynomial for the Alvarez lenses consists of only three coefficients. By computing spatially varying point spread functions (PSFs), we simulate the degraded images of multiple zoom configurations and conduct restoration experiments. The results demonstrate that our approach surpasses the compared methods in the restoration of degraded images across various zoom configurations while also exhibiting strong generalization capabilities under untrained configurations.

Response to Singh et al. 2023: It is premature for a unified hypothesis of sudden unexpected death in epilepsy: A great amount of research is still needed to understand the multisystem cascade.

In response to the comments by Singh and colleagues about our recent paper proposing a unified hypothesis of SUDEP, we definitely agree that more research is needed. This research should include studies in other models, including Dravet mice, emphasized by Singh et al. However, we strongly believe the hypothesis is timely, because it is based on the continuing progress on SUDEP-related research on serotonin (5-HT) and adenosine as well as neuroanatomical findings.We propose testing of 5-HT enhancing drugs, neurotoxicity blocking drugs, such as N-methyl-D-aspartate (NMDA) antagonists and periaqueductal gray (PAG) electrical stimulation for SUDEP prevention. There are FDA-approved drugs that enhance the action of 5-HT, including fluoxetine and fenfluramine, which is approved for Dravet syndrome. NMDA antagonists, including memantine and ketamine, are also approved for other disorders. PAG electrical stimulation, which is proposed to activate a suffocation alarm, is also approved to treat other conditions and is known to enhance respiration. Experiments using these methods are currently being carried out in animal studies. If these approaches are validated in SUDEP models, treatments could be evaluated relatively quickly in patients with epilepsy (PWE) who exhibit a biomarker for high SUDEP risk, such as peri-ictal respiratory abnormalities. An example of such a study is the ongoing clinical trial of a selective serotonin reuptake inhibitor in PWE. Although, gene-based therapies may ultimately become treatments of choice to prevent SUDEP, as Singh et al suggested, one or more of the approaches we proposed could become temporizing treatments before gene-based therapies can be available. Establishing genetic treatments would require extensive time for each of the genetic abnormalities associated with SUDEP, and too many PWE are likely to die in the meantime.The temporizing treatments may help to reduce the incidence of SUDEP sooner, which is urgently needed.

A unified hypothesis of SUDEP: Seizure-induced respiratory depression induced by adenosine may lead to SUDEP but can be prevented by autoresuscitation and other restorative respiratory response mechanisms mediated by the action of serotonin on the periaqueductal gray.

Sudden unexpected death in epilepsy (SUDEP) is a major cause of death in people with epilepsy (PWE). Postictal apnea leading to cardiac arrest is the most common sequence of terminal events in witnessed cases of SUDEP, and postconvulsive central apnea has been proposed as a potential biomarker of SUDEP susceptibility. Research in SUDEP animal models has led to the serotonin and adenosine hypotheses of SUDEP. These neurotransmitters influence respiration, seizures, and lethality in animal models of SUDEP, and are implicated in human SUDEP cases. Adenosine released during seizures is proposed to be an important seizure termination mechanism. However, adenosine also depresses respiration, and this effect is mediated, in part, by inhibition of neuronal activity in subcortical structures that modulate respiration, including the periaqueductal gray (PAG). Drugs that enhance the action of adenosine increase postictal death in SUDEP models. Serotonin is also released during seizures, but enhances respiration in response to an elevated carbon dioxide level, which often occurs postictally. This effect of serotonin can potentially compensate, in part, for the adenosine-mediated respiratory depression, acting to facilitate autoresuscitation and other restorative respiratory response mechanisms. A number of drugs that enhance the action of serotonin prevent postictal death in several SUDEP models and reduce postictal respiratory depression in PWE. This effect of serotonergic drugs may be mediated, in part, by actions on brainstem sites that modulate respiration, including the PAG. Enhanced activity in the PAG increases respiration in response to hypoxia and other exigent conditions and can be activated by electrical stimulation. Thus, we propose the unifying hypothesis that seizure-induced adenosine release leads to respiratory depression. This can be reversed by serotonergic action on autoresuscitation and other restorative respiratory responses acting, in part, via the PAG. Therefore, we hypothesize that serotonergic or direct activation of this brainstem site may be a useful approach for SUDEP prevention.

Long non-coding RNA HOTTIP exerts an oncogenic function by regulating HOXA13 in nasopharyngeal carcinoma.

BACKGROUND: The long non-coding RNA HOXA transcript at the distal tip (HOTTIP) and homeobox A13 (HOXA13) have been identified as oncogenes that play a pivotal role in tumorigenesis. However, their specific mechanisms of action in nasopharyngeal carcinoma (NPC) progression remain unclear. METHODS AND RESULTS: In the present study, RT-qPCR was employed to quantify RNA expression in NPC cells and tissues. Flow cytometry, MTT, CCK8 and colony formation assays were utilized to assess cell apoptosis and proliferation. Transwell assay was conducted to evaluate migration and invasion while Western blotting was performed for protein expression analysis. Our findings revealed that the expression of HOTTIP was significantly upregulated in NPC cell lines. Inhibition of HOTTIP could induce apoptosis and suppress proliferation, clonogenicity, invasion and metastasis in NPC cells. Knockdown of HOTTIP led to downregulation of HOXA13 expression, which subsequently inhibited the proliferation and metastasis in NPC cells. The inhibitory effects on cell proliferation and metastasis caused by HOTTIP silencing were rescued by HOXA13 overexpression. Additionally, there was a significant positive correlation between HOTTIP and HOXA13, which were found to be elevated in NPC tissues compared to normal tissues. CONCLUSIONS: We have determined that LncRNA HOTTIP facilitates tumorigenesis by modulating the expression of HOXA13 in NPC cells. Targeting HOTTIP/HOXA13 may be a promising therapeutic strategy for NPC.

Energy-efficient electrochemical treatment of paracetamol using a PbO2 anode based on pulse electrodeposition strategy: Kinetics, energy consumption and mechanism.

The purpose of this research is to study the pulse electrochemical oxidation of paracetamol (PCT) using a novel PbO2 anode based on pulse electrodeposition strategy (PbO2-PE). The pulse electrodeposition strategy used to prepare a PbO2 anode resulted in rougher surface, higher directional specificity of ß(101) and more redox couples of Pb4+/Pb2+. Additionally, the oxygen evolution potential (OEP) and charge transfer resistance were also improved. When compared to direct current electrochemical oxidation process, pulse electrolysis in had a slightly higher PCT removal efficiency and active species (·OH and active chlorine) production, while 72.04% of energy consumption was saved. The effects of operating parameters on PCT degradation efficiency and specific energy consumption were studied. The findings suggested that the pulse electrochemical oxidation of PCT followed a pseudo-first-order kinetic model, with PCT removal reaching 98.63% after 60 min of electrolysis under optimal conditions. Possible mechanisms describing reaction pathways for PCT were also proposed. Finally, combinating with the economic feasibility and safety evaluation, we could conclude that pulse electrolysis with a PbO2-PE electrode was a promising option for improving the practicability of electrochemical treatment for refractory organic wastewater.

Mechanism of highly efficient electrochemical degradation of antibiotic sulfadiazine using a layer-by-layer GNPs/PbO2 electrode.

A PbO2 electrode integrating electrocatalytic and adsorptive functions was successfully fabricated by embedding layer-by-layer graphene nanoplatelets (GNPs) into ß-PbO2 active layer (GNPs/PbO2) and employed as anode for high-efficient removal of sulfadiazine (SDZ). In electrochemical degradation experiments, SDZ was quickly enriched on the surface of GNPs/PbO2 film via adsorption and then oxidized by ⋅OH in-site. In terms of the electrocatalytic performance and adsorption of electrode, the optimal electrodeposition time for each ß-PbO2 outer layer was 4 min (GNPs/PbO2-4). Compared with conventional PbO2 electrode, the layer-by-layer GNPs resulted in the smaller crystal size and denser surface of PbO2 electrode, thus facilitating the generation of active oxygen species. At the same time, the specific surface area, oxygen evolution potential (OEP) of the anode were enhanced and the charge-transfer resistance was reduced. For GNPs/PbO2-4 anode, the optimal conditions of electrochemical oxidation of SDZ were identified as initial pH 9, 50 mg/L of SDZ and 20 mA/cm2 of current density using response surface methodology (RSM), 98.15% of SDZ could be removed in this case. The contribution of radical oxidation and non-radical oxidation to SDZ removal was about 79% and 21%, respectively. Moreover, the reaction pathways of SDZ on the GNPs/PbO2-4 electrode involving hydroxylation, radical reaction and ring cleavage were speculated. Finally, the continuous SDZ degradation and accelerated service lifetime test suggested that the GNPs/PbO2-4 electrode was shown to be stable and repeatable, and the Pb2+ concentration was measured to ensure the safety of the treated solution. Consequently, the above findings provide an innovative way to design and prepare an effective and stable PbO2 electrode for electrochemical degradation of antibiotic wastewater.

Direct distortion prediction method for AR-HUD dynamic distortion correction.

Dynamic distortion is one of the most critical factors affecting the experience of automotive augmented reality head-up displays (AR-HUDs). A wide range of views and the extensive display area result in extraordinarily complex distortions. Existing methods based on the neural network first obtain distorted images and then get the predistorted data for training mostly. This paper proposes a distortion prediction framework based on the neural network. It directly trains the network with the distorted data, realizing dynamic adaptation for AR-HUD distortion correction and avoiding errors in coordinate interpolation. Additionally, we predict the distortion offsets instead of the distortion coordinates and present a field of view (FOV)-weighted loss function based on the spatial-variance characteristic to further improve the prediction accuracy of distortion. Experiments show that our methods improve the prediction accuracy of AR-HUD dynamic distortion without increasing the network complexity or data processing overhead.

Design of an optimized Alvarez lens based on the fifth-order polynomial combination.

This paper proposes an optimized design of the Alvarez lens by utilizing a combination of three fifth-order X-Y polynomials. It can effectively minimize the curvature of the lens surface to meet the manufacturing requirements. The phase modulation function and aberration of the proposed lens are evaluated by using first-order optical analysis. Simulations compare the proposed lens with the traditional Alvarez lens in terms of surface curvature, zoom capability, and imaging quality. The results demonstrate the exceptional performance of the proposed lens, achieving a remarkable 26.36% reduction in the maximum curvature of the Alvarez lens (with a coefficient A value of 4×10-4 and a diameter of 26 mm) while preserving its original zoom capability and imaging quality.

Non-blind optical degradation correction via frequency self-adaptive and finetune tactics.

In mobile photography applications, limited volume constraints the diversity of optical design. In addition to the narrow space, the deviations introduced in mass production cause random bias to the real camera. In consequence, these factors introduce spatially varying aberration and stochastic degradation into the physical formation of an image. Many existing methods obtain excellent performance on one specific device but are not able to quickly adapt to mass production. To address this issue, we propose a frequency self-adaptive model to restore realistic features of the latent image. The restoration is mainly performed in the Fourier domain and two attention mechanisms are introduced to match the feature between Fourier and spatial domain. Our method applies a lightweight network, without requiring modification when the fields of view (FoV) changes. Considering the manufacturing deviations of a specific camera, we first pre-train a simulation-based model, then finetune it with additional manufacturing error, which greatly decreases the time and computational overhead consumption in implementation. Extensive results verify the promising applications of our technique for being integrated with the existing post-processing systems.