Supervised learning algorithms generally assume the availability of enough memory to store data models during the training and test phases. However, this assumption is unrealistic when data comes in the form of infini...
详细信息
Although supervised image denoising networks have shown remarkable performance on synthesized noisy images, they often fail in practice due to the difference between real and synthesized noise. Since clean-noisy image...
Although supervised image denoising networks have shown remarkable performance on synthesized noisy images, they often fail in practice due to the difference between real and synthesized noise. Since clean-noisy image pairs from the real world are extremely costly to gather, self-supervised learning, which utilizes noisy input itself as a target, has been studied. To prevent a self-supervised denoising model from learning identical mapping, each output pixel should not be influenced by its corresponding input pixel; This requirement is known as J-invariance. Blind-spot networks (BSNs) have been a prevalent choice to ensure J-invariance in self-supervised image denoising. However, constructing variations of BSNs by injecting additional operations such as downsampling can expose blinded information, thereby violating J-invariance. Consequently, convolutions designed specifically for BSNs have been allowed only, limiting architectural flexibility. To overcome this limitation, we propose PUCA, a novel J-invariant U-Net architecture, for self-supervised denoising. PUCA leverages patch-unshuffle/shuffle to dramatically expand receptive fields while maintaining J-invariance and dilated attention blocks (DABs) for global context incorporation. Experimental results demonstrate that PUCA achieves state-of-the-art performance, outperforming existing methods in self-supervised image denoising.
The advancement of communication technologies and cloud systems has led to the emergence of the Healthcare-Consumer Internet of Things (H-CIoT) as a significant domain. This emergence has transformed the traditional h...
详细信息
Modern applications can generate a large amount of data from different sources with high velocity, a combination that is difficult to store and process via traditional tools. Hadoop is one framework that is used for t...
详细信息
With the increasing use of solar energy, DC Microgrid with lower capacities are formed with connection to the traditional main grid. If a short circuit fault occurs in DC microgrids, the contribution of multiple sourc...
详细信息
ISBN:
(数字)9798350331592
ISBN:
(纸本)9798350331608
With the increasing use of solar energy, DC Microgrid with lower capacities are formed with connection to the traditional main grid. If a short circuit fault occurs in DC microgrids, the contribution of multiple sources including main grid and DC sources might lead to very high fault levels that are difficult to break. While technologies for manufacturing the DC circuit breaker with higher breaking capacity are on-going, another approach by changing tripping sequences in DC protection systems, which could reduce fault currents to be interrupted by a single circuit breaker, was studied in this paper. With the aid of MATLAB/Simulink and OPAL-RT 5700 real time simulation, effectiveness of the proposed approach was evaluated on a Simulink stimulation model of 500 V, 1 MWgrid-connected DC Microgrid. The result obtained indicated that the fault current level interrupted by the DC circuit breakers was reduced by up to 63%.
In this paper, we explore the relationship between an individual’s writing style and the risk that they will engage in online harmful behaviors (such as cyberbullying). In particular, we consider whether measurable d...
详细信息
Few-shot learning (FSL) is the process of rapid generalization from abundant base samples to inadequate novel samples. Despite extensive research in recent years, FSL is still not yet able to generate satisfactory sol...
详细信息
Water scarcity critically threatens agricultural sus-tainability in Jordan, necessitating innovative technological solutions. This study presents a comprehensive framework that integrates Internet of Things (IoT) sens...
详细信息
ISBN:
(数字)9798331523657
ISBN:
(纸本)9798331523664
Water scarcity critically threatens agricultural sus-tainability in Jordan, necessitating innovative technological solutions. This study presents a comprehensive framework that integrates Internet of Things (IoT) sensors with Big Data Analytics to optimize water usage in agriculture. Employing a multi-stage approach, we conducted real-time data acquisition from diverse farms, processed the data using clustering algorithms, and implemented intelligent decision-making systems. Our analysis focused on key parameters such as soil moisture, temperature, rainfall, and wind speed across various farming locations. The methodology encompassed extensive field data collection, advanced analytics, and the deployment of smart irrigation systems. Results demonstrated a significant 25 % reduction in water consumption while maintaining optimal crop yields. Additionally, clustering identified three distinct farm categories, each requiring tailored water management strategies. This research contributes to the field of smart agriculture by providing practical, scalable solutions for regions facing severe water scarcity. The findings highlight the potential of advanced technologies to enhance agricultural sustainability and address critical water management challenges.
This paper proposes an analytical target modifi-cation for linear robust model predictive control strategies in order to deal with time-varying references defined by dynamic signal targets. The new approach can be dir...
详细信息
ISBN:
(数字)9798350382655
ISBN:
(纸本)9798350382662
This paper proposes an analytical target modifi-cation for linear robust model predictive control strategies in order to deal with time-varying references defined by dynamic signal targets. The new approach can be directly integrated to linear robust model predictive control algorithms that achieve piecewise constant reference tracking if recursive feasibility is ensured for any set-point. The main contribution is to present a direct analytical approach that provides a potentially improved steady-state tracking error performance with the same computation complexity of the original MPC for tracking piecewise constant reference. A simulation case study based on the trajectory tracking control of a quadrotor is used to illustrate the usefulness of the new analytical target modification layer.
VA3, is a novel web3.0 based individual peer-to-peer platform for electricity settlement between individual New Zealanders with the ability to trade at a sub-household level i.e., multiple accounts within a single hom...
详细信息
ISBN:
(数字)9798350331592
ISBN:
(纸本)9798350331608
VA3, is a novel web3.0 based individual peer-to-peer platform for electricity settlement between individual New Zealanders with the ability to trade at a sub-household level i.e., multiple accounts within a single home or a portable account per person. We were able create a web3 application based on the base Ethereum network to successfully automate at individual person’s power consumption and production which is fed into a smart contract using a software update to modify the home router to basically act as a home energy management system and then settlements are automatically managed by the smart contracts.
暂无评论