In this paper, we propose a maximum a posteriori ramework for the super-resolution problem, i.e., reconstructing high-resolution images from shifted, rotated, low-resolution degraded observations. The main contributions of this work are two; first, the use of a new locally adaptive edge preserving prior for the super-resolution problem. Second an efficient two-step reconstruction methodology that includes first an initial registration using only the low-resolution degraded observations. This is followed by a fast iterative algorithm implemented in the discrete Fourier transform domain in which the restoration, interpolation and the registration subtasks of this problem are preformed simultaneously. We present examples with both synthetic and real data that demonstrate the advantages of the proposed framework.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/tip.2007.896664 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
GAN-based data reconstruction attacks in split learning.
Neural Netw
January 2025
Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan 430072, China. Electronic address:
Due to the distinctive distributed privacy-preserving architecture, split learning has found widespread application in scenarios where computational resources on the client side are limited. Unlike clients in federated learning retaining the whole model, split learning partitions the model into two segments situated separately on the server and client ends, thereby preventing direct access to the complete model structure by either party and fortifying its resilience against attacks. However, existing studies have demonstrated that even with access restricted to partial model outputs, split learning remains susceptible to data reconstruction attacks.
View Article and Find Full Text PDFIdentifying the Optimal Sampling Strategy for the Bayesian Estimation of Vancomycin AUC in Adult Hematologic Cancer Patients.
Clin Pharmacokinet
January 2025
Faculté de Pharmacie, Université de Montréal, Montréal, QC, Canada.
Background And Objective: The latest consensus recommends using the ratio between the area under the curve over 24 h (AUC) and minimal inhibitory concentration (MIC) as the therapeutic target for vancomycin in clinical practice, with a Bayesian approach and population pharmacokinetic (popPK) model being particularly recommended. While using both post-dose peak concentration (C) and pre-dose concentration (C) is more accurate than C alone, the optimal sampling strategy for estimating AUC is still unclear. The objective of this study was to determine the best sampling time(s) to estimate AUC using the Bayesian approach in these specific adult hematologic cancer patients.
View Article and Find Full Text PDFIn this Letter, we propose a high-performance optimized detection scheme based on a neural network (NN) in a receiver digital signal processing (DSP) for bandwidth-limited intensity modulation and direct detection (IM/DD) transmission systems. The NN-based optimized detection scheme consists of two components, an NN-based lookup table (NN-LUT) and an NN-based log-maximum estimation with a fixed number of surviving state (NN-MAP) decoder. The NN-LUT provides more accurate and sufficient information (PI) to the decoder than the conventional filter-form PI without increasing computational complexity.
View Article and Find Full Text PDFImpaired arbitration between reward-related decision-making strategies in Alcohol Users compared to Alcohol Non-Users: a computational modeling study.
NPP Digit Psychiatry Neurosci
January 2025
Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, NY USA.
Article Synopsis
- Reinforcement learning contrasts model-free (habitual) and model-based (goal-directed) decision-making, with the latter being more common in high-reward scenarios.
- A study involving 81 participants revealed that frequent alcohol users lacked the ability to adjust their decision-making strategies based on reward levels, unlike non-users who displayed better model-based control in high-reward settings.
- Both groups were less risk-averse in high stakes, but alcohol users were generally more prone to risky decisions and showed impaired flexibility in adapting to changing reward conditions.
Comparison and calibration of MP2RAGE quantitative T1 values to multi-TI inversion recovery T1 values.
Magn Reson Imaging
January 2025
Department of Electrical and Computer Engineering, Vanderbilt University, Nashville, TN, United States; Department of Computer Science, Vanderbilt University, Nashville, TN, United States; Vanderbilt University Institute of Imaging Science, Vanderbilt University Medical Center, Nashville, TN, United States; Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, United States; Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, United States.
While typical qualitative T1-weighted magnetic resonance images reflect scanner and protocol differences, quantitative T1 mapping aims to measure T1 independent of these effects. Changes in T1 in the brain reflect structural changes in brain tissue. Magnetization-prepared two rapid acquisition gradient echo (MP2RAGE) is an acquisition protocol that allows for efficient T1 mapping with a much lower scan time per slab compared to multi-TI inversion recovery (IR) protocols.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!