Denoising(DN) and demosaicing(DM) are the first crucial stages in the image signal processing pipeline. Recently, researches pay more attention to solve DN and DM in a joint manner, which is an extremely undetermined ...
详细信息
Denoising(DN) and demosaicing(DM) are the first crucial stages in the image signal processing pipeline. Recently, researches pay more attention to solve DN and DM in a joint manner, which is an extremely undetermined inverse problem. Existing deep learning methods learn the desired prior on synthetic dataset, which limits the generalization of learned network to the real world data. Moreover, existing methods mainly focus on the raw data property of high green information sampling rate for DM, but occasionally exploit the high intensity and signalto-noise(SNR) of green channel. In this work, a deep guided attention network(DGAN) is presented for real image joint DN and DM(JDD), which considers both high SNR and high sampling rate of green information for DN and DM, respectively. To ease the training and fully exploit the data property of green channel, we first train DN and DM sub-networks sequentially and then learn them jointly, which can alleviate the error accumulation. Besides, in order to support the real image JDD, we collect paired raw clean RGB and noisy mosaic images to conduct a realistic dataset. The experimental results on real JDD dataset show the presented approach performs better than the state-of-the-art methods, in terms of both quantitative metrics and qualitative visualization.
The virtual private cloud service currently lacks a real-time end-to-end consistency validation mechanism, which prevents tenants from receiving immediate feedback on their requests. Existing solutions consume excessi...
详细信息
The virtual private cloud service currently lacks a real-time end-to-end consistency validation mechanism, which prevents tenants from receiving immediate feedback on their requests. Existing solutions consume excessive communication and computational resources in such large-scale cloud environments, and suffer from poor timeliness. To address these issues, we propose a lightweight consistency validation mechanism that includes real-time incremental validation and periodic full-scale validation. The former leverages message layer aggregation to enable tenants to swiftly determine the success of their requests on hosts with minimal communication overhead. The latter utilizes lightweight validation checksums to compare the expected and actual states of hosts locally, while efficiently managing the checksums of various host entries using inverted indexing. This approach enables us to efficiently validate the complete local configurations within the limited memory of hosts. In summary, our proposed mechanism achieves closed-loop implementation for new requests and ensures their long-term effectiveness.
BiFeO_(3)(BFO)has received considerable attention as a lead-free ferroelectric film due to its large theoretical remnant ***,BFO suffers from a large leakage current,resulting in poor ferroelectric ***,the sol-gel met...
详细信息
BiFeO_(3)(BFO)has received considerable attention as a lead-free ferroelectric film due to its large theoretical remnant ***,BFO suffers from a large leakage current,resulting in poor ferroelectric ***,the sol-gel method was used to deposit a series of BFO-based thin films on fluorine-doped tin oxide substrates,and the effects of the substitution of the elements Co,Cu,Mn(B-site)and Sm,Eu,La(A-site)on the crystal structure,ferroelectricity,and leakage current of the BFO-based thin films were *** confirmed that lattice distortion by X-ray diffraction can be attributed to the substitution of individual elements in the BFO-based *** and Eu substitutions contribute to the lattice distortion in a pseudo-cubic structure,while La is biased toward *** force microscopy confirmed that reversible switching of ferroelectric domains by nearly 180°can be realized through the prepared *** ferroelectric hysteresis loops showed that the order for the polarization contribution is as follows:Cu>Co>Mn(B-site),Sm>La>Eu(A-site).The current density voltage curves indicated that the order for leakage contribution is as follows:Mn
An Al2O3 dispersion strengthened(ADS)alloy with an ultra-high softening temperature of∼1200 K was fabricated by the in-situ internal oxidation and reduction *** evolution of the nanometer Al2O3 particles,grain size,a...
详细信息
An Al2O3 dispersion strengthened(ADS)alloy with an ultra-high softening temperature of∼1200 K was fabricated by the in-situ internal oxidation and reduction *** evolution of the nanometer Al2O3 particles,grain size,and consequently the softening behavior of this ADS alloy,were investigated by conducting the annealing treatments in the range from 673 K to 1273 K for 60 *** refined nanometer Al2O3 particles were found to be highly stable at elevated temperatures,leading to the high dislocation density and grain boundary stability of the *** average grain size was found to increase extremely slowly from∼0.60μm to∼0.74μm with increasing annealing temperatures from 773 K to 1273 K.A criterion for grain boundaries migration and softening was established based on the competition between grain growth and pinning effect of Al2O3 *** strong pinning effect of Al2O3 particles was found when the grain size was between the lower limit(about 0.4-0.5μm)and upper limit(2.18μm).The occurrence of softening behavior was attributed to the rapid increase of the proportion of grains larger than the upper limit.A modified Hall-Petch relationship was established by introducing the integration of the grain size distribution,which can describe this correlation between softening behavior and the pinning effect of Al2O3 *** current study not only sheds light on the further understanding of the softening mechanism of ADS copper alloy but also provides a useful route for designing copper alloy with high softening resistance.
Flapping-Wing Air Vehicles(FWAVs)have been developed to pursue the efficient,agile,and quiet flight of flying ***,unlike lightweight FWAVs capable of vertical takeoff,relatively heavy FWAVs face challenges in self-tak...
详细信息
Flapping-Wing Air Vehicles(FWAVs)have been developed to pursue the efficient,agile,and quiet flight of flying ***,unlike lightweight FWAVs capable of vertical takeoff,relatively heavy FWAVs face challenges in self-takeoff,which refers to taking off without both external device and energy *** this study,a cliff-drop method is implemented for an independent takeoff of a heavy FWAV,relying solely on *** the takeoff process using the cliff-drop method,the FWAV moves on the ground to a cliff edge using a wheel-driving motor and then descends from the cliff to achieve the necessary speed for *** demonstrate the cliff-drop method,the KAIST Robotic Hawk(KRoHawk)with a mass of 740 g and a wingspan of 120 cm is *** takeoff tests demonstrate that the KRoHawk,significantly heavier than the vertical-takeoff capable FWAVs,can successfully take off using the gravity-assisted takeoff *** scalability of cliff-drop method is analyzed through *** drop constraints are absent,the wheel-driving motor mass fraction for cliff-drop method remains negligible even as the vehicle's weight *** drop constraints are set to 4 m,FWAVs heavier than KRoHawk,weighing up to 4.4 kg,can perform the cliff-drop takeoffs with a wheel-driving motor mass fraction of less than 8%.
Accurate mechanical system models are crucial for safe and stable control. Unlike linear systems,Lagrangian systems are highly nonlinear and difficult to optimize because of their unknown system *** research thus used...
详细信息
Accurate mechanical system models are crucial for safe and stable control. Unlike linear systems,Lagrangian systems are highly nonlinear and difficult to optimize because of their unknown system *** research thus used deep neural networks to generate linear models of original systems by mapping nonlinear dynamic systems into a linear space with a Koopman observable function encoder. The controller then relies on the Koopman linear model. However, without physical information constraints, ensuring control consistency between the original nonlinear system and the Koopman system is tough, as the learning process of the Koopman observation function is unsupervised. This paper thus proposes a two-stage learning algorithm that uses structural subnetworks to build a physics-informed network topology to simultaneously learn the Koopman observable functions and the system energy representation. In the Koopman matrix learning session, a quadratic-constrained optimization problem is solved to ensure that the Koopman representation satisfies the energy difference matching hard constraint. The proposed energy-preserving deep Lagrangian Koopman(EPDLK) framework effectively represents the dynamics of the Lagrangian system while ensuring control consistency. The effectiveness of EPDLK is compared with those of various Koopman observable function construction methods in multistep prediction and trajectory tracking tasks. EPDLK achieves better control consistency by guaranteeing energy difference matching, which facilitates the application of the control law generated on the Koopman system directly to the original nonlinear Lagrangian system.
Weathering steel exhibits excellent corrosion resistance and is widely used in bridges,towers,railways,highways,and other engineering projects that are exposed to the atmosphere for long periods of ***,before the form...
详细信息
Weathering steel exhibits excellent corrosion resistance and is widely used in bridges,towers,railways,highways,and other engineering projects that are exposed to the atmosphere for long periods of ***,before the formation of stable rust layers,weathering steel is prone to liquid rust sagging and spattering,leading to environmental pollution and city appearance *** factors limit the application and development of weathering *** this study,a rapid and environmentally friendly method was de-veloped by introducing alloying elements,specifically investigating the role of Sn in the rapid stabilization of rust layers in marine atmo-spheric *** rust layer formed on weathering low-alloy steel exposed to prolonged outdoor conditions and laboratory im-mersion experiments was explored using electron probe micro-analyzer(EPMA),micro-Raman,X-ray photoelectron spectroscopy(XPS),and electrochemical *** showed an optimal synergistic effect between Sn and Cr,which facilitated the accelerated densification of the rust *** beneficial effect enhanced the capability of the rust layer to resist Cl^(-)erosion and improved the protec-tion performance of the rust layer.
For the rational manipulation of the production quality of high-temperature metallurgical engineering,there are many challenges in understanding the processes involved because of the black box chemical/electrochemical...
详细信息
For the rational manipulation of the production quality of high-temperature metallurgical engineering,there are many challenges in understanding the processes involved because of the black box chemical/electrochemical *** overcome this issue,various in-situ characterization methods have been recently developed to analyze the interactions between the composition,microstructure,and solid-liquid interface of high-temperature electrochemical electrodes and molten *** this review,recent progress of in-situ hightemperature characterization techniques is discussed to summarize the advances in understanding the processes in metallurgical ***-situ high-temperature technologies and analytical methods mainly include synchrotron X-ray diffraction(s-XRD),laser scanning confocal microscopy,and X-ray computed microtomography(X-rayμ-CT),which are important platforms for analyzing the structure and morphology of the electrodes to reveal the complexity and variability of their *** addition,laser-induced breakdown spectroscopy,high-temperature Raman spectroscopy,and ultraviolet-visible absorption spectroscopy provide microscale characterizations of the composition and structure of molten *** importantly,the combination of X-rayμ-CT and s-XRD techniques enables the investigation of the chemical reaction mechanisms at the two-phase ***,these in-situ methods are essential for analyzing the chemical/electrochemical kinetics of high-temperature reaction processes and establishing the theoretical principles for the efficient and stable operation of chemical/electrochemical metallurgical processes.
The ever-growing energy demand and environmental issues have stimulated the development of sustainable energy ***,an efficient and environmentally friendly electrochemical transformation technology was proposed to pre...
详细信息
The ever-growing energy demand and environmental issues have stimulated the development of sustainable energy ***,an efficient and environmentally friendly electrochemical transformation technology was proposed to prepare highly graphitized carbon materials from an abundant natural resource-lignin (LG).The preparation process mainly includes pyrolytic carbonization of raw LG material and electrochemical conversion of amorphous carbon ***,with the assistance of Co catalyst,the graphitization degree of the products was significantly improved,in which the mechanism was the removal of heteroatoms in LG and the rearrangement of carbon atoms into graphite ***,tunable microstructures (nanoflakes) under catalytic effects could also be observed by controlling the electrolytic *** with the products CN1 (without catalyst) and CN5 (with 10%catalyst),the specific surface area are 158.957 and 202.246 m^(2)g^(-1),*** used as the electrode material for lithium-ion batteries,CN5 delivered a competitive specific capacity of~350 m Ah g^(-1)(0.5 C) compared with commercial *** strategy proposed in this work provides an effective way to extract value-added graphite materials from lignin and can be extended to the graphitization conversion of any other amorphous carbon precursor materials.
Augmenting the working voltage is an effective way to maximize the energy density of Ni-rich layered Li[Ni0.8Co0.1Mn0.1]O2(NCM) to approach its theoretical ***,NCM suffers from structural degradation in deep delithi...
详细信息
Augmenting the working voltage is an effective way to maximize the energy density of Ni-rich layered Li[Ni0.8Co0.1Mn0.1]O2(NCM) to approach its theoretical ***,NCM suffers from structural degradation in deep delithiation state,which is often accompanied by severe surface lattice oxygen loss and transition metal dissolution,leading to restricted cycle ***,a facile and effective surfacestrengthening strategy is proposed,in which Mn(OH)2nanoshells are uniformly grown on the NCM surface as a Li+capturer and then converted to thin spinel Li4Mn5O12layers during subsequent hightemperature *** resultant Li4Mn5O12layers can enhance cathode-electrolyte interface electrochemical stability with inhibited electrolyte corrosion and accelerated Li+*** theoretical calculations confirms that the Mn-O bonds formed at the interfaces can effectively decrease the oxygen activity,thereby further inhibiting the lattice oxygen release and structural degradation caused by the irreversible phase ***,the Li4Mn5O12-coated NCM displays high capacity retention of 80.3% and 94.9% at 1 C and 5 C compared to the pristine NCM(52.5% and 10.1%) after 200 cycles and can operate stably at 2.7-4.6 V and 60℃.The spinel Li4Mn5O12-coating demonstrates an effective route to enhance the structural/electrochemical stability of NCM for next-generation advanced lithium-ion batteries.
暂无评论