Articles de revues sur le sujet « Unrolling methods »
Pour voir les autres types de publications sur ce sujet consultez le lien suivant : Unrolling methods.
Créez une référence correcte selon les styles APA, MLA, Chicago, Harvard et plusieurs autres
Consultez les 34 meilleurs articles de revues pour votre recherche sur le sujet « Unrolling methods ».
À côté de chaque source dans la liste de références il y a un bouton « Ajouter à la bibliographie ». Cliquez sur ce bouton, et nous générerons automatiquement la référence bibliographique pour la source choisie selon votre style de citation préféré : APA, MLA, Harvard, Vancouver, Chicago, etc.
Vous pouvez aussi télécharger le texte intégral de la publication scolaire au format pdf et consulter son résumé en ligne lorsque ces informations sont inclues dans les métadonnées.
Parcourez les articles de revues sur diverses disciplines et organisez correctement votre bibliographie.
1
Lee, Songil, Gyouhyung Kyung, Minjoong Kim, Donghee Choi, Hyeeun Choi, Kitae Hwang, Seonghyeok Park, Su Young Kim et Seungbae Lee. « Shaping Rollable Display Devices : Effects of Gripping Condition, Device Thickness, and Hand Length on Bimanual Perceived Grip Comfort ». Human Factors : The Journal of the Human Factors and Ergonomics Society 62, no 5 (2 juillet 2019) : 770–86. http://dx.doi.org/10.1177/0018720819855225.
Texte intégralRésumé :
Objective To examine the effects of the gripping condition, device thickness, and hand length on bimanual perceived grip comfort associated with unrolling hand-held rollable screens. Background Rollable displays can be rolled and unrolled to change screen size. Although diverse rollable display device concepts have been suggested, little is known regarding ergonomic forms for comfortable screen unrolling. Method Thirty young individuals (10 in each hand-length group) evaluated three rollable display device prototypes in three gripping conditions (no restriction on using side bezels, minimal use of side bezels, and restriction on the gripping type). Prototypes differed in their right-side thickness (2, 6, and 10 mm). Side bezel regions grasped during screen unrolling and corresponding bimanual grip comfort ratings were obtained. Results To improve perceived grip comfort and accommodate user-preferred gripping methods, rollable display devices should be 6 mm (preferably 10 mm) thick (vs. 2 mm) and have at least 20-mm-wide side bezels. Relative to device thickness, gripping conditions were more influential on grip comfort ratings. The “no restriction” condition improved grip comfort ratings and strengthened bimanual coupling in terms of grip comfort ratings. Conclusion Contrary to current smartphone trends toward thinner and bezel-less designs, hand-held rollable display devices should be sufficiently thick and have sufficiently wide side bezels for screen unrolling. Application Hand-held rollable display devices should be 6- or preferably 10-mm thick (vs. 2 mm) and have at least 20-mm-wide side bezels to ensure higher perceived grip comfort during bilateral screen unrolling.
2
Song, Heping, Qifeng Ding, Jingyao Gong, Hongying Meng et Yuping Lai. « SALSA-Net : Explainable Deep Unrolling Networks for Compressed Sensing ». Sensors 23, no 11 (28 mai 2023) : 5142. http://dx.doi.org/10.3390/s23115142.
Texte intégralRésumé :
Deep unrolling networks (DUNs) have emerged as a promising approach for solving compressed sensing (CS) problems due to their superior explainability, speed, and performance compared to classical deep network models. However, the CS performance in terms of efficiency and accuracy remains a principal challenge for approaching further improvements. In this paper, we propose a novel deep unrolling model, SALSA-Net, to solve the image CS problem. The network architecture of SALSA-Net is inspired by unrolling and truncating the split augmented Lagrangian shrinkage algorithm (SALSA) which is used to solve sparsity-induced CS reconstruction problems. SALSA-Net inherits the interpretability of the SALSA algorithm while incorporating the learning ability and fast reconstruction speed of deep neural networks. By converting the SALSA algorithm into a deep network structure, SALSA-Net consists of a gradient update module, a threshold denoising module, and an auxiliary update module. All parameters, including the shrinkage thresholds and gradient steps, are optimized through end-to-end learning and are subject to forward constraints to ensure faster convergence. Furthermore, we introduce learned sampling to replace traditional sampling methods so that the sampling matrix can better preserve the feature information of the original signal and improve sampling efficiency. Experimental results demonstrate that SALSA-Net achieves significant reconstruction performance compared to state-of-the-art methods while inheriting the advantages of explainable recovery and high speed from the DUNs paradigm.
3
Yu, Youhao, et Richard M. Dansereau. « MsDC-DEQ-Net : Deep Equilibrium Model (DEQ) with Multiscale Dilated Convolution for Image Compressive Sensing (CS) ». IET Signal Processing 2024 (18 janvier 2024) : 1–12. http://dx.doi.org/10.1049/2024/6666549.
Texte intégralRésumé :
Compressive sensing (CS) is a technique that enables the recovery of sparse signals using fewer measurements than traditional sampling methods. To address the computational challenges of CS reconstruction, our objective is to develop an interpretable and concise neural network model for reconstructing natural images using CS. We achieve this by mapping one step of the iterative shrinkage thresholding algorithm (ISTA) to a deep network block, representing one iteration of ISTA. To enhance learning ability and incorporate structural diversity, we integrate aggregated residual transformations (ResNeXt) and squeeze-and-excitation mechanisms into the ISTA block. This block serves as a deep equilibrium layer connected to a semi-tensor product network for convenient sampling and providing an initial reconstruction. The resulting model, called MsDC-DEQ-Net, exhibits competitive performance compared to state-of-the-art network-based methods. It significantly reduces storage requirements compared to deep unrolling methods, using only one iteration block instead of multiple iterations. Unlike deep unrolling models, MsDC-DEQ-Net can be iteratively used, gradually improving reconstruction accuracy while considering computation tradeoffs. Additionally, the model benefits from multiscale dilated convolutions, further enhancing performance.
4
Zhang, Linrui, Qin Zhang, Li Shen, Bo Yuan, Xueqian Wang et Dacheng Tao. « Evaluating Model-Free Reinforcement Learning toward Safety-Critical Tasks ». Proceedings of the AAAI Conference on Artificial Intelligence 37, no 12 (26 juin 2023) : 15313–21. http://dx.doi.org/10.1609/aaai.v37i12.26786.
Texte intégralRésumé :
Safety comes first in many real-world applications involving autonomous agents. Despite a large number of reinforcement learning (RL) methods focusing on safety-critical tasks, there is still a lack of high-quality evaluation of those algorithms that adheres to safety constraints at each decision step under complex and unknown dynamics. In this paper, we revisit prior work in this scope from the perspective of state-wise safe RL and categorize them as projection-based, recovery-based, and optimization-based approaches, respectively. Furthermore, we propose Unrolling Safety Layer (USL), a joint method that combines safety optimization and safety projection. This novel technique explicitly enforces hard constraints via the deep unrolling architecture and enjoys structural advantages in navigating the trade-off between reward improvement and constraint satisfaction. To facilitate further research in this area, we reproduce related algorithms in a unified pipeline and incorporate them into SafeRL-Kit, a toolkit that provides off-the-shelf interfaces and evaluation utilities for safety-critical tasks. We then perform a comparative study of the involved algorithms on six benchmarks ranging from robotic control to autonomous driving. The empirical results provide an insight into their applicability and robustness in learning zero-cost-return policies without task-dependent handcrafting. The project page is available at https://sites.google.com/view/saferlkit.
5
Kamali, Hadi Mardani, et Shaahin Hessabi. « A Fault Tolerant Parallelism Approach for Implementing High-Throughput Pipelined Advanced Encryption Standard ». Journal of Circuits, Systems and Computers 25, no 09 (21 juin 2016) : 1650113. http://dx.doi.org/10.1142/s0218126616501139.
Texte intégralRésumé :
Advanced Encryption Standard (AES) is the most popular symmetric encryption method, which encrypts streams of data by using symmetric keys. The current preferable AES architectures employ effective methods to achieve two important goals: protection against power analysis attacks and high-throughput. Based on a different architectural point of view, we implement a particular parallel architecture for the latter goal, which is capable of implementing a more efficient pipelining in field-programmable gate array (FPGA). In this regard, all intermediate registers which have a role for unrolling the main loop will be removed. Also, instead of unrolling the main loop of AES algorithm, we implement pipelining structure by replicating nonpipelined AES architectures and using an auto-assigner mechanism for each AES block. By implementing the new pipelined architecture, we achieve two valuable advantages: (a) solving single point of failure problem when one of the replicated parts is faulty and (b) deploying the proposed design as a fault tolerant AES architecture. In addition, we put emphasis on area optimization for all four AES main functions to reduce the overhead associated with AES block replication. The simulation results show that the maximum frequency of our proposed AES architecture is 675.62[Formula: see text]MHz, and for AES128 the throughput is 86.5[Formula: see text]Gbps which is 30.9% better than its closest existing competitor.
6
Мельник, Л. М., А. С. Конотоп et О. П. Кизимчук. « ЗАСТОСУВАННЯ ТРАДИЦІЙНИХ НАЦІОНАЛЬНИХ ЕЛЕМЕНТІВ ОЗДОБЛЕННЯ В СУЧАСНОМУ ОДЯЗІ ». Art and Design, no 2 (15 juin 2018) : 51–58. http://dx.doi.org/10.30857/2617-0272.2018.2.6.
Texte intégralRésumé :
The purpose of this work is to establish the possibility of the decorating of modern women's clothes with artistic elements and ornaments of traditional national clothes of Ukraine different regions and their subsequent reproduction in knitwear. Methods of literary-analytical review and visual-analytical research are used.The elements of clothing decoration that are characteristic of ethnic Ukrainian clothing have been defined during the research. The structures of knitted fabrics based on on openwork, plated and interlooping with unrolling, which simulate different merezhka of traditional Ukrainian embroidery, have been developed in this study. The few variants of simple mesh are offered: narrow one made on the basis of the rib 1 + 1 with an unrolling; narrow complex canvas - on the basis of openwork interlooping; wide (more than 2 cm) canvas - on the basis of a plated interlooping. The model of the women's dress is executed in the semi-regular way on two-bar flat-knitted machine, using the developed variants of knitted fabric structures. Taking into account the traditions of an arrangement of the embroidery decoration elements and modern fashion trends, it is proposed to use a narrow simple snap-net to decorate the neck of the product, a wide snap-net, having a plant ornament to knit 2/3 of the bottom of the sleeve. The narrow canvas is also used as a connecting element of the sleeve parts made with various interlooping. The decoration elements of national Ukrainian clothes have been investigated and reproduced in women's knitted dress by new knitted structure creation taking into account modern technology. Research results can be used to expand the range of modern women's clothing
7
Ye, Yutong, Hongyin Zhu, Chaoying Zhang et Binghai Wen. « Efficient graphic processing unit implementation of the chemical-potential multiphase lattice Boltzmann method ». International Journal of High Performance Computing Applications 35, no 1 (27 octobre 2020) : 78–96. http://dx.doi.org/10.1177/1094342020968272.
Texte intégralRésumé :
The chemical-potential multiphase lattice Boltzmann method (CP-LBM) has the advantages of satisfying the thermodynamic consistency and Galilean invariance, and it realizes a very large density ratio and easily expresses the surface wettability. Compared with the traditional central difference scheme, the CP-LBM uses the Thomas algorithm to calculate the differences in the multiphase simulations, which significantly improves the calculation accuracy but increases the calculation complexity. In this study, we designed and implemented a parallel algorithm for the chemical-potential model on a graphic processing unit (GPU). Several strategies were used to optimize the GPU algorithm, such as coalesced access, instruction throughput, thread organization, memory access, and loop unrolling. Compared with dual-Xeon 5117 CPU server, our methods achieved 95 times speedup on an NVIDIA RTX 2080Ti GPU and 106 times speedup on an NVIDIA Tesla P100 GPU. When the algorithm was extended to the environment with dual NVIDIA Tesla P100 GPUs, 189 times speedup was achieved and the workload of each GPU reached 96%.
8
Guo, Yang, Wei Gao, Siwei Ma et Ge Li. « Accelerating Transform Algorithm Implementation for Efficient Intra Coding of 8K UHD Videos ». ACM Transactions on Multimedia Computing, Communications, and Applications 18, no 4 (30 novembre 2022) : 1–20. http://dx.doi.org/10.1145/3507970.
Texte intégralRésumé :
Real-time ultra-high-definition (UHD) video applications have attracted much attention, where the encoder side urgently demands the high-throughput two-dimensional (2D) transform hardware implementation for the latest video coding standards. This article proposes an effective acceleration method for transform algorithm in UHD intra coding based on the third generation of audio video coding standard (AVS3). First, by conducting detailed statistical analysis, we devise an efficient hardware-friendly transform algorithm that can reduce running cycles and resource consumption remarkably. Second, to implement multiplierless computation for saving resources and power, a series of shift-and-add unit (SAU) hardwares are investigated to have much less adoptions of shifters and adders than the existing methods. Third, different types of hardware acceleration methods, including calculation pipelining, logical-loop unrolling, and module-level parallelism, are designed to efficaciously support the data-intensive high frame-rate 8K UHD video coding. Finally, due to the scarcity of 8K video sources, we also provide a new dataset for the performance verification. Experimental results demonstrate that our proposed method can effectively fulfill the real-time 8K intra encoding at beyond 60 fps, with very negligible loss on rate-distortion (R-D) performance, which is averagely 0.98% Bjontegaard-Delta Bit-Rate (BD-BR).
9
Aydin, Seda Guzel, et Hasan Şakir Bilge. « FPGA Implementation of Image Registration Using Accelerated CNN ». Sensors 23, no 14 (21 juillet 2023) : 6590. http://dx.doi.org/10.3390/s23146590.
Texte intégralRésumé :
Background: Accurate and fast image registration (IR) is critical during surgical interventions where the ultrasound (US) modality is used for image-guided intervention. Convolutional neural network (CNN)-based IR methods have resulted in applications that respond faster than traditional iterative IR methods. However, general-purpose processors are unable to operate at the maximum speed possible for real-time CNN algorithms. Due to its reconfigurable structure and low power consumption, the field programmable gate array (FPGA) has gained prominence for accelerating the inference phase of CNN applications. Methods: This study proposes an FPGA-based ultrasound IR CNN (FUIR-CNN) to regress three rigid registration parameters from image pairs. To speed up the estimation process, the proposed design makes use of fixed-point data and parallel operations carried out by unrolling and pipelining techniques. Experiments were performed on three US datasets in real time using the xc7z020, and the xcku5p was also used during implementation. Results: The FUIR-CNN produced results for the inference phase 139 times faster than the software-based network while retaining a negligible drop in regression performance of under 200 MHz clock frequency. Conclusions: Comprehensive experimental results demonstrate that the proposed end-to-end FPGA-based accelerated CNN achieves a negligible loss, a high speed for registration parameters, less power when compared to the CPU, and the potential for real-time medical imaging.
10
Wang, Nan, Xiaoling Zhang, Tianwen Zhang, Liming Pu, Xu Zhan, Xiaowo Xu, Yunqiao Hu, Jun Shi et Shunjun Wei. « A Sparse-Model-Driven Network for Efficient and High-Accuracy InSAR Phase Filtering ». Remote Sensing 14, no 11 (30 mai 2022) : 2614. http://dx.doi.org/10.3390/rs14112614.
Texte intégralRésumé :
Phase filtering is a vital step for interferometric synthetic aperture radar (InSAR) terrain elevation measurements. Existing phase filtering methods can be divided into two categories: traditional model-based and deep learning (DL)-based. Previous studies have shown that DL-based methods are frequently superior to traditional ones. However, most of the existing DL-based methods are purely data-driven and neglect the filtering model, so that they often need to use a large-scale complex architecture to fit the huge training sets. The issue brings a challenge to improve the accuracy of interferometric phase filtering without sacrificing speed. Therefore, we propose a sparse-model-driven network (SMD-Net) for efficient and high-accuracy InSAR phase filtering by unrolling the sparse regularization (SR) algorithm to solve the filtering model into a network. Unlike the existing DL-based filtering methods, the SMD-Net models the physical process of filtering in the network and contains fewer layers and parameters. It is thus expected to ensure the accuracy of the filtering without sacrificing speed. In addition, unlike the traditional SR algorithm setting the spare transform by handcrafting, a convolutional neural network (CNN) module was established to adaptively learn such a transform, which significantly improved the filtering performance. Extensive experimental results on the simulated and measured data demonstrated that the proposed method outperformed several advanced InSAR phase filtering methods in both accuracy and speed. In addition, to verify the filtering performance of the proposed method under small training samples, the training samples were reduced to 10%. The results show that the performance of the proposed method was comparable on the simulated data and superior on the real data compared with another DL-based method, which demonstrates that our method is not constrained by the requirement of a huge number of training samples.
11
Ariño-Gutierrez, Mayte, Mercedes Molero-Senosiain, Barbara Burgos-Blasco, Beatriz Vidal-Villegas, Pedro Arriola-Villalobos, Jose Antonio Gegundez-Fernandez, Gregory Moloney et Luis Daniel Holguín. « Challenges of DMEK Technique with Young Corneal Donors’ Grafts : Surgical Keys for Success—A Pilot Study ». Journal of Clinical Medicine 12, no 19 (30 septembre 2023) : 6316. http://dx.doi.org/10.3390/jcm12196316.
Texte intégralRésumé :
Purpose: To report on the surgical maneuvers recommended for a successful unfolding of very young donors in order to accomplish an uneventful Descemet Membrane Endothelial Keratoplasty (DMEK) surgery. Methods: Five patients (three females and two males, mean age 71.2 ± 6.7 years) with Fuchs endothelial cell dystrophy who underwent DMEK with very young donors (between 20 and 30 years old) were included. The following demographic data were assessed: donor’s age, donor’s endothelial cell density (ECD), preservation time, recipient’s age and sex and unfolding surgical time. Best-corrected visual acuity (BCVA; decimal system), ECD and corneal central thickness (CCT) were assessed preoperatively and at 6-month follow-up. Results: Donors’ mean age was 23.6 ± 3.6 years (range 21 to 30) and the mean ECD was 2748.6 ± 162.6 cells/mm2. All of them underwent an uneventful DMEK as a single procedure performed by one experienced surgeon (MAG) with a mean unfolding time of 7.2 ± 4.9 min (range 4 to 15). The essential steps, including patient preparation as well as DMEK graft implantation, orientation, unrolling and centering are detailed. At 6 months, BCVA was 0.6 ± 0.2, ECD was 1945.0 ± 455.5 cells/mm2 and CCT was 497.0 ± 19.7 microns. Conclusions: We hereby present the keys to overcome tightly scrolled grafts of very young donors, which prove perfectly suitable for DMEK surgery. The graft shape tends towards a double-roll and specific maneuvers are strongly recommended.
12
Liang, Yong, Junwen Tan, Zhisong Xie, Zetao Chen, Daoqian Lin et Zhenhao Yang. « Research on Convolutional Neural Network Inference Acceleration and Performance Optimization for Edge Intelligence ». Sensors 24, no 1 (31 décembre 2023) : 240. http://dx.doi.org/10.3390/s24010240.
Texte intégralRésumé :
In recent years, edge intelligence (EI) has emerged, combining edge computing with AI, and specifically deep learning, to run AI algorithms directly on edge devices. In practical applications, EI faces challenges related to computational power, power consumption, size, and cost, with the primary challenge being the trade-off between computational power and power consumption. This has rendered traditional computing platforms unsustainable, making heterogeneous parallel computing platforms a crucial pathway for implementing EI. In our research, we leveraged the Xilinx Zynq 7000 heterogeneous computing platform, employed high-level synthesis (HLS) for design, and implemented two different accelerators for LeNet-5 using loop unrolling and pipelining optimization techniques. The experimental results show that when running at a clock speed of 100 MHz, the PIPELINE accelerator, compared to the UNROLL accelerator, experiences an 8.09% increase in power consumption but speeds up by 14.972 times, making the PIPELINE accelerator superior in performance. Compared to the CPU, the PIPELINE accelerator reduces power consumption by 91.37% and speeds up by 70.387 times, while compared to the GPU, it reduces power consumption by 93.35%. This study provides two different optimization schemes for edge intelligence applications through design and experimentation and demonstrates the impact of different quantization methods on FPGA resource consumption. These experimental results can provide a reference for practical applications, thereby providing a reference hardware acceleration scheme for edge intelligence applications.
13
Lam, Duc Khai, Cam Vinh Du et Hoai Luan Pham. « QuantLaneNet : A 640-FPS and 34-GOPS/W FPGA-Based CNN Accelerator for Lane Detection ». Sensors 23, no 15 (25 juillet 2023) : 6661. http://dx.doi.org/10.3390/s23156661.
Texte intégralRésumé :
Lane detection is one of the most fundamental problems in the rapidly developing field of autonomous vehicles. With the dramatic growth of deep learning in recent years, many models have achieved a high accuracy for this task. However, most existing deep-learning methods for lane detection face two main problems. First, most early studies usually follow a segmentation approach, which requires much post-processing to extract the necessary geometric information about the lane lines. Second, many models fail to reach real-time speed due to the high complexity of model architecture. To offer a solution to these problems, this paper proposes a lightweight convolutional neural network that requires only two small arrays for minimum post-processing, instead of segmentation maps for the task of lane detection. This proposed network utilizes a simple lane representation format for its output. The proposed model can achieve 93.53% accuracy on the TuSimple dataset. A hardware accelerator is proposed and implemented on the Virtex-7 VC707 FPGA platform to optimize processing time and power consumption. Several techniques, including data quantization to reduce data width down to 8-bit, exploring various loop-unrolling strategies for different convolution layers, and pipelined computation across layers, are optimized in the proposed hardware accelerator architecture. This implementation can process at 640 FPS while consuming only 10.309 W, equating to a computation throughput of 345.6 GOPS and energy efficiency of 33.52 GOPS/W.
14
Wang, Ling, Hai Zhou, Chunjiang Bian, Kangning Jiang et Xiaolei Cheng. « Hardware Acceleration and Implementation of YOLOX-s for On-Orbit FPGA ». Electronics 11, no 21 (26 octobre 2022) : 3473. http://dx.doi.org/10.3390/electronics11213473.
Texte intégralRésumé :
The rapid development of remote sensing technology has brought about a sharp increase in the amount of remote sensing image data. However, due to the satellite’s limited hardware resources, space, and power consumption constraints, it is difficult to process massive remote sensing images efficiently and robustly using the traditional remote sensing image processing methods. Additionally, the task of satellite-to-ground target detection has higher requirements for speed and accuracy under the conditions of more and more remote sensing data. To solve these problems, this paper proposes an extremely efficient and reliable acceleration architecture for forward inference of the YOLOX-s detection network an on-orbit FPGA. Considering the limited onboard resources, the design strategy of the parallel loop unrolling of the input channels and output channels is adopted to build the largest DSP computing array to ensure a reliable and full utilization of the limited computing resources, thus reducing the inference delay of the entire network. Meanwhile, a three-path cache queue and a small-scale cascaded pooling array are designed, which maximize the reuse of on-chip cache data, effectively reduce the bandwidth bottleneck of the external memory, and ensure an efficient computing of the entire computing array. The experimental results show that at the 200 MHz operating frequency of the VC709, the overall inference performance of the FPGA acceleration can reach 399.62 GOPS, the peak performance can reach 408.4 GOPS, and the overall computing efficiency of the DSP array can reach 97.56%. Compared with the previous work, our architecture design further improves the computing efficiency under limited hardware resources.
15
Buglaev, Anatoly M. « Device for Wood-Cutting Tool Hardening ». Lesnoy Zhurnal (Forestry Journal), no 5 (15 octobre 2021) : 134–41. http://dx.doi.org/10.37482/0536-1036-2021-5-134-141.
Texte intégralRésumé :
Choosing effective methods and devices for surface hardening of wood-cutting tools is problematic due to the variety of their designs and operating conditions. In this regard, the development of such devices becomes an urgent task. According to the literature, one of the effective methods for increasing the service life of machine parts and tools is electrospark hardening or electrospark alloying. Industrial electrospark installations such as “EFI” (electrophysical measurements) and “Elitron” with manual vibrators are used for electrospark hardening. However, using manual vibrators significantly increases the labour intensity and hardening time. Moreover, the surface quality after hardening with manual vibrators is often unsatisfactory. Various mechanized installations have been developed in order to reduce the labour intensity of electrospark hardening. Nevertheless, these installations are designed to harden specific parts and do not allow hardening tools of various designs, including woodcutting tools. The surface quality after hardening in mechanized installations does not always satisfy the customer. Further surface plastic deformation treatments, such as rolling and unrolling with rollers and balls, as well as diamond burnishing, are often used to improve the surface quality after electrospark hardening. The surface quality after additional processing by these methods boosts, although the labour intensity and cost of the hardening process increase. To increase the wear resistance of machine parts and tools, it is reasonable to reduce the height parameters of roughness, increase microhardness, and form the residual compressive stresses, which is ensured by the methods of surface plastic deformation. In this regard, it becomes necessary to use electrospark hardening simultaneously with surface plastic deformation. The work presents the design and features of using the device for hardening. The device was used to strengthen the thicknesser machine knives, which made it possible to almost double their durability. Applying this device, in comparison with using the electrospark hardening with a manual vibrator, reduces the roughness of the hardened surface and improves the surface quality of the processed workpieces. The modes of hardening have been installed, making it possible to effectively harden wood-cutting tools. For citation: Buglaev A.M. Device for Wood-Cutting Tool Hardening. Lesnoy Zhurnal [Russian Forestry Journal], 2021, no. 5, pp. 134–141. DOI: 10.37482/0536-1036-2021-5-134-141
16
Lai, Zeqiang, Kaixuan Wei, Ying Fu, Philipp Härtel et Felix Heide. « ∇-Prox : Differentiable Proximal Algorithm Modeling for Large-Scale Optimization ». ACM Transactions on Graphics 42, no 4 (26 juillet 2023) : 1–19. http://dx.doi.org/10.1145/3592144.
Texte intégralRésumé :
Tasks across diverse application domains can be posed as large-scale optimization problems, these include graphics, vision, machine learning, imaging, health, scheduling, planning, and energy system forecasting. Independently of the application domain, proximal algorithms have emerged as a formal optimization method that successfully solves a wide array of existing problems, often exploiting problem-specific structures in the optimization. Although model-based formal optimization provides a principled approach to problem modeling with convergence guarantees, at first glance, this seems to be at odds with black-box deep learning methods. A recent line of work shows that, when combined with learning-based ingredients, model-based optimization methods are effective, interpretable, and allow for generalization to a wide spectrum of applications with little or no extra training data. However, experimenting with such hybrid approaches for different tasks by hand requires domain expertise in both proximal optimization and deep learning, which is often error-prone and time-consuming. Moreover, naively unrolling these iterative methods produces lengthy compute graphs, which when differentiated via autograd techniques results in exploding memory consumption, making batch-based training challenging. In this work, we introduce ∇-Prox, a domain-specific modeling language and compiler for large-scale optimization problems using differentiable proximal algorithms. ∇-Prox allows users to specify optimization objective functions of unknowns concisely at a high level, and intelligently compiles the problem into compute and memory-efficient differentiable solvers. One of the core features of ∇-Prox is its full differentiability, which supports hybrid model- and learning-based solvers integrating proximal optimization with neural network pipelines. Example applications of this methodology include learning-based priors and/or sample-dependent inner-loop optimization schedulers, learned with deep equilibrium learning or deep reinforcement learning. With a few lines of code, we show ∇-Prox can generate performant solvers for a range of image optimization problems, including end-to-end computational optics, image deraining, and compressive magnetic resonance imaging. We also demonstrate ∇-Prox can be used in a completely orthogonal application domain of energy system planning, an essential task in the energy crisis and the clean energy transition, where it outperforms state-of-the-art CVXPY and commercial Gurobi solvers.
17
Popov, Valery A., et Valentin V. Elantsev. « On the improvement efficiency and safety of operation of underground tunnel escalators. Adaptation of processes of planning and control ». Izvestiya MGTU MAMI 17, no 3 (25 décembre 2023) : 305–19. http://dx.doi.org/10.17816/2074-0530-492274.
Texte intégralRésumé :
BACKGROUND: The pace of transport system development not corresponding to current demands of a town is a significant obstacle on the way of achieving target values of business activity and quality of life. One of the main indicators of quality of life in a modern large city is satisfaction with city public transport. The backbone mean of public transport in Saint Petersburg is underground which efficiency indicators are share of accomplishment of train schedule and accomplishment of stations’ operation schedule. However, funding based on subsidies and requirement of proper fulfillment of given efficiency indicators make the Petersburg Underground SUE look for new ways of adaptation of planning and control processes for managing available resources ensuring keeping outdating infrastructure operable and cost reduction based on state-of-the-art informational technologies.
AIMS: Establishment of enlarged hardware facility and basic structural solution necessary for unrolling software for planning and control of maintenance of the underground’s escalator park. The study object is the underground’s escalator park; the study subject is automation of maintenance system.
METHODS: In the first part of the paper, as a result of statistical compilation and comparison of homogeneous data from annual reports of the Petersburg Underground SUE in the tabular form as well as analysis of qualitative and quantitative indicators of the escalator park considering the information on the history of maintenance system formation, the statement was made about existence of interaction between the changes taking place in the city transport system and demands of adaptation of processes of planning and control of operations in the escalator park maintenance system. In the second part of the paper, on the basis of analysis of regulatory framework and requirements to development of data processing centers, the main structural element for implementation of facility of planning and control of performing the underground’s escalator park maintenance was established.
RESULTS: The theoretically valuable result of the study is the found interaction between changes taking place at the level of the city transport system and demands in changing the underground’s escalator park maintenance system for achieving the comprehensive evolution of the transport system. Meanwhile, the basic structure of the hardware-software facility, main structural element of which is the corridor isolation system implementing the tool of adaptation of processes of planning and control the operation, is significantly valuable for practical use.
CONCLUSIONS: Structure and equipment of the hardware-software facility as well as the installed software for monitoring, planning and control of the operations ensuring/not ensuring fulfillment the efficiency indicators give the opportunity to transfer the infrastructure parts from the running hours-based maintenance system to the state-based maintenance system.
18
Park, BoSun, et Seog Chung Seo. « Efficient Implementation of NIST LWC ESTATE Algorithm Using OpenCL and Web Assembly for Secure Communication in Edge Computing Environment ». Sensors 21, no 6 (11 mars 2021) : 1987. http://dx.doi.org/10.3390/s21061987.
Texte intégralRésumé :
In edge computing service, edge devices collect data from a number of embedded devices, like sensors, CCTVs (Closed-circuit Television), and so on, and communicate with application servers. Since a large portion of communication in edge computing services are conducted in wireless, the transmitted data needs to be properly encrypted. Furthermore, the application servers (resp. edge devices) are responsible for encrypting or decrypting a large amount of data from edge devices (resp. terminal devices), the cryptographic operation needs to be optimized on both server side and edge device side. Actually, the confidentiality and integrity of data are essential for secure communication. In this paper, we present two versions of security software which can be used on edge device side and server side for secure communication between them in edge computing environment. Our softwares are basically web-based application because of its universality where the softwares can be executed on any web browsers. Our softwares make use of ESTATE (Energy efficient and Single-state Tweakable block cipher based MAC-Then-Encrypt)algorithm, which is a promising candidate of NIST LWC (National Institute of Standards and Technology LightWeight Cryptography) competition and it provides not only data confidentiality but also data authentication. It also implements the ESTATE algorithm using Web Assembly for efficient use on edge devices, and optimizes the performance of the algorithm using the properties of the underlying block cipher. Several methods are applied to efficiently operate the ESTATE algorithm. We use conditional statements to XOR the extended tweak values during the operation of the ESTATE algorithm. To eliminate this unnecessary process, we use a method of expanding and storing the tweak value through pre-computation. The measured results of the ESTATE algorithm implemented with Web Assembly and the reference C/C++ ESTATE algorithm are compared. ESTATE implemented as Web Assembly is measured in web browsers Chrome, FireFox, and Microsoft Edge. For efficiency on server side, we make use of OpenCL which is parallel computing framework in order to process a number of data simultaneously. In addition, when implementing with OpenCL, using conditional statements causes performance degradation. We eliminated the conditional statement using the loop unrolling method to eliminate the performance degradation. In addition, OpenCL operates by moving the data to be encrypted to the local memory because the local memory has a high operation speed. TweAES-128 and TweAES-128-6, which have the same structure as AES algorithm, can apply the previously existing studied T-table method. In addition, the input value 16-byte is processed in parallel and calculated. In addition, since it may be vulnerable to cache-timing attack, it is safely operated by applying the previously existing studied T-table shuffling method. Our softwares cover the necessary security service from edge devices to servers in edge computing services and they can be easily used for various types of edge computing devices because they are all web-based applications.
19
Liu, Shuangli, Pengcheng Wan et Xin Shang. « HASR-TAI : Hybrid model-based interpretable network and super-resolution network for thermoacoustic imaging ». Applied Physics Letters 123, no 13 (25 septembre 2023). http://dx.doi.org/10.1063/5.0169109.
Texte intégralRésumé :
Microwave induced thermoacoustic tomography has shown promise for noninvasive and non-ionizing early tumor detection. Nowadays, thermoacoustic reconstruction methods based on deep learning have achieved good and time-efficient results. However, both deep learning methods based on the initial thermoacoustic image and end-to-end methods lack interpretability due to the black-box property of neural networks. In this Letter, we propose an interpretable end-to-end network structure comprising an unrolling part and a super-resolution part. In the unrolling part, a deep unfolding network interprets the iterations of the model-based algorithm based on compressed sensing as layers of the network. Subsequently, a fast and efficient super-resolution neural network maps the low-resolution image into the super-resolution space. Two breast models with different sizes of tumor targets are used for validation. By comparing with the traditional method and the deep learning method, the proposed method demonstrates superior performance in image quality and imaging time. Moreover, the parameters in the network hold physical significance, offering the potential for the interpretable end-to-end network in thermoacoustic imaging.
20
Shi, Baoshun, Ke Jiang, Shaolei Zhang, Qiusheng Lian, Yanwei Qin et Yunsong Zhao. « Mud-Net : Multi-domain deep unrolling network for simultaneous sparse-view and metal artifact reduction in computed tomography ». Machine Learning : Science and Technology, 5 janvier 2024. http://dx.doi.org/10.1088/2632-2153/ad1b8e.
Texte intégralRésumé :
Abstract Sparse-view computed tomography (SVCT) is regarded as a promising technique to accelerate data acquisition and reduce radiation dose. However, in the presence of metallic implants, SVCT inevitably makes the reconstructed CT images suffer from severe metal artifacts and streaking artifacts due to the lack of sufficient projection data. Previous stand-alone SVCT and metal artifact reduction (MAR) methods to solve the problem of simultaneously sparse-view and metal artifact reduction (SVMAR) are plagued by insufficient correction accuracy. To overcome this limitation, we propose a multi-domain deep unrolling network, called Mud-Net, for SVMAR. Specifically, we establish a joint sinogram, image, artifact, and coding domains deep unrolling reconstruction model to recover high-quality CT images from the under-sampled sinograms corrupted by metallic implants. To train this multi-domain network effectively, we embed multi-domain knowledge into the network training process. Comprehensive experiments demonstrate that our method is superior to both existing MAR methods in the full-view MAR task and previous SVCT methods in the SVMAR task.
21
Solod, Panadda, Nattha Jindapetch, Kiattisak Sengchuai, Apidet Booranawong, Pakpoom Hoyingcharoen, Surachate Chumpol et Masami Ikura. « High Level Synthesis Optimizations of Road Lane Detection Development on Zynq-7000 ». Pertanika Journal of Science and Technology 29, no 2 (30 avril 2021). http://dx.doi.org/10.47836/pjst.29.2.01.
Texte intégralRésumé :
In this work, we proposed High-Level Synthesis (HLS) optimization processes to improve the speed and the resource usage of complex algorithms, especially nested-loop. The proposed HLS optimization processes are divided into four steps: array sizing is performed to decrease the resource usage on Programmable Logic (PL) part, loop analysis is performed to determine which loop must be loop unrolling or loop pipelining, array partitioning is performed to resolve the bottleneck of loop unrolling and loop pipelining, and HLS interface is performed to select the best block level and port level interface for array argument of RTL design. A case study road lane detection was analyzed and applied with suitable optimization techniques to implement on the Xilinx Zynq-7000 family (Zybo ZC7010-1) which was a low-cost FPGA. From the experimental results, our proposed method reaches 6.66 times faster than the primitive method at clock frequency 100 MHz or about 6 FPS. Although the proposed methods cannot reach the standard real-time (25 FPS), they can instruct HLS developers for speed increasing and resource decreasing on an FPGA.
22
Je, Changsoo, et Hyung-Min Park. « Binary Stripe Unwrapping Based on Mean-Speed Walk and Local Median Correction for Rapid High-Resolution Structured-Light Range Imaging ». International Journal of Sensors, Wireless Communications and Control 13 (19 décembre 2022). http://dx.doi.org/10.2174/2210327913666221219091440.
Texte intégralRésumé :
Aim: Structured light is frequently selected for efficient and accurate depth imaging, and single-frame-based methods have been presented for real-time sensing or imaging dynamic objects. However, many existing single-frame-based methods do not provide sufficient range resolution. Even those capable of sufficient range resolution mostly result in insufficient signal-to-noise ratio or depend on spatially windowed uniqueness, where a larger window makes the identification trickier. Method: This paper presents a novel method for rapid structured-light range sensing using a binary color stripe pattern. For accurate and reliable depth acquisition, we identify projected stripes by our stripe segmentation and unwrapping algorithms. For robust stripe detection, the color-stripe segmentation algorithm performs image upsizing, motion blurring, and color balancing. The binary stripe unwrapping algorithm consists of mean-speed walk unrolling, row-wise unrolling, and local median correction, and resolves the high-frequency color-stripe redundancy efficiently and reliably. objective: We provide a novel method for rapid structured-light range sensing using a binary color stripe pattern. Result: Experimental results show the effectiveness and reliability of the presented method. Conclusion: Even using an entry-level phone camera under a low-cost DLP projector produces high-accuracy results.
23
Zhou, Qingping, Jaiyu Qian, Junqi Tang et Jinglai Li. « Deep unrolling networks with recurrent momentum acceleration for nonlinear inverse problems ». Inverse Problems, 20 mars 2024. http://dx.doi.org/10.1088/1361-6420/ad35e3.
Texte intégralRésumé :
Abstract Combining the strengths of model-based iterative algorithms and data-driven deep learning solutions, deep unrolling networks (DuNets) have become a popular tool to solve inverse imaging problems. While DuNets have been successfully applied to many linear inverse problems, nonlinear problems tend to impair the performance of the method. Inspired by momentum acceleration techniques that are often used in optimization algorithms, we propose a recurrent momentum acceleration (RMA) framework that uses a long short-term memory recurrent neural network (LSTM-RNN) to simulate the momentum acceleration process. The RMA module leverages the ability of the LSTM-RNN to learn and retain knowledge from the previous gradients. We apply RMA to two popular DuNets -- the learned proximal gradient descent (LPGD) and the learned primal-dual (LPD) methods, resulting in LPGD-RMA and LPD-RMA respectively. We provide experimental results on two nonlinear inverse problems: a nonlinear deconvolution problem, and an electrical impedance tomography problem with limited boundary measurements. In the first experiment we have observed that the improvement due to RMA largely increases with respect to the nonlinearity of the problem. The results of the second example further demonstrate that the RMA schemes can significantly improve the performance of DuNets in strongly ill-posed problems.
24
Li, Tianao, et E. Alexander. « Galaxy image deconvolution for weak gravitational lensing with unrolled plug-and-play ADMM ». Monthly Notices of the Royal Astronomical Society : Letters, 10 mars 2023. http://dx.doi.org/10.1093/mnrasl/slad032.
Texte intégralRésumé :
Abstract Removing optical and atmospheric blur from galaxy images significantly improves galaxy shape measurements for weak gravitational lensing and galaxy evolution studies. This ill-posed linear inverse problem is usually solved with deconvolution algorithms enhanced by regularisation priors or deep learning. We introduce a so-called ”physics-informed deep learning” approach to the Point Spread Function (PSF) deconvolution problem in galaxy surveys. We apply algorithm unrolling and the Plug-and-Play technique to the Alternating Direction Method of Multipliers (ADMM), in which a neural network learns appropriate hyperparameters and denoising priors from simulated galaxy images. We characterise the time-performance trade-off of several methods for galaxies of differing brightness levels as well as our method’s robustness to systematic PSF errors and network ablations. We show an improvement in reduced shear ellipticity error of 38.6% (SNR=20)/45.0% (SNR=200) compared to classic methods and 7.4% (SNR=20)/33.2% (SNR=200) compared to modern methods.
25
Gu, Hongyi, Burhaneddin Yaman, Steen Moeller, Jutta Ellermann, Kamil Ugurbil et Mehmet Akçakaya. « Revisiting ℓ1-wavelet compressed-sensing MRI in the era of deep learning ». Proceedings of the National Academy of Sciences 119, no 33 (8 août 2022). http://dx.doi.org/10.1073/pnas.2201062119.
Texte intégralRésumé :
Following their success in numerous imaging and computer vision applications, deep-learning (DL) techniques have emerged as one of the most prominent strategies for accelerated MRI reconstruction. These methods have been shown to outperform conventional regularized methods based on compressed sensing (CS). However, in most comparisons, CS is implemented with two or three hand-tuned parameters, while DL methods enjoy a plethora of advanced data science tools. In this work, we revisit ℓ 1 -wavelet CS reconstruction using these modern tools. Using ideas such as algorithm unrolling and advanced optimization methods over large databases that DL algorithms utilize, along with conventional insights from wavelet representations and CS theory, we show that ℓ 1 -wavelet CS can be fine-tuned to a level close to DL reconstruction for accelerated MRI. The optimized ℓ 1 -wavelet CS method uses only 128 parameters compared to >500,000 for DL, employs a convex reconstruction at inference time, and performs within <1% of a DL approach that has been used in multiple studies in terms of quantitative quality metrics.
26
Szécsi, Péter György, Gábor Horváth et Zoltán Porkoláb. « Improved Loop Execution Modeling in the Clang Static Analyzer ». Acta Cybernetica, 22 octobre 2020. http://dx.doi.org/10.14232/actacyb.283176.
Texte intégralRésumé :
The LLVM Clang Static Analyzer is a source code analysis tool which aims to find bugs in C, C++, and Objective-C programs using symbolic execution, i.e. it simulates the possible execution paths of the code. Currently the simulation of the loops is somewhat naive (but efficient), unrolling the loops a predefined constant number of times. However, this approach can result in a loss of coverage in various cases. This study aims to introduce two alternative approaches which can extend the current method and can be applied simultaneously: (1) determining loops worth to fully unroll with applied heuristics, and (2) using a widening mechanism to simulate an arbitrary number of iteration steps. These methods were evaluated on numerous open source projects, and proved to increase coverage in most of the cases. This work also laid the infrastructure for future loop modeling improvements.
27
Moos, Thorben. « Unrolled Cryptography on Silicon ». IACR Transactions on Cryptographic Hardware and Embedded Systems, 26 août 2020, 416–42. http://dx.doi.org/10.46586/tches.v2020.i4.416-442.
Texte intégralRésumé :
Cryptographic primitives with low-latency performance have gained momentum lately due to an increased demand for real-time applications. Block ciphers such as PRINCE enable data encryption (resp. decryption) within a single clock cycle at a moderately high operating frequency when implemented in a fully-unrolled fashion. Unsurprisingly, many typical environments for unrolled ciphers require protection against physical adversaries as well. Yet, recent works suggest that most common SCA countermeasures are hard to apply to low-latency circuits. Hardware masking, for example, requires register stages to offer resistance, thus adding delay and defeating the purpose of unrolling. On another note, it has been indicated that unrolled primitives without any additional means of protection offer an intrinsic resistance to SCA attacks due to their parallelism, asynchronicity and speed of execution. In this work, we take a closer look at the physical security properties provided by unrolled cryptographic IC implementations. We are able to confirm that the nature of unrolling indeed bears the potential to decrease the susceptibility of cipher implementations significantly when reset methods are applied. With respect to certain adversarial models, e.g., ciphertext-only access, an amazingly high level of protection can be achieved. While this seems to be a great result for cryptographic hardware engineers, there is an attack vector hidden in plain sight which still threatens the security of unrolled implementations remarkably – namely the static power consumption of CMOS-based circuits. We point out that essentially all reasons which make it hard to extract meaningful information from the dynamic behavior of unrolled primitives are not an issue when exploiting the static currents for key recovery. Our evaluation is based on real-silicon measurements of an unrolled PRINCE core in a custom 40nm ASIC. The presented results serve as a neat educational case study to demonstrate the broad differences between dynamic and static power information leakage in the light of technological advancement.
28
Rajmohan, Shathanaa, N. Ramasubramanian et Nagi Naganathan. « Design Space Exploration for Reducing Cost of Hardware Trojan Detection and Isolation during Architectural Synthesis ». Journal of Circuits, Systems and Computers, 29 décembre 2020, 2150156. http://dx.doi.org/10.1142/s0218126621501565.
Texte intégralRésumé :
In past years, software used to be the main concern of computer security, and the hardware was assumed to be safe. However, Hardware Trojans, which are a malicious alteration to the circuit, pose a threat to the security of a system. Trojans may be distributed across different components of the system and can bring down the security by communicating with each other. Redundancy and vendor diversity-based methods exist to detect Hardware Trojans, but with an increase in the hardware overhead. This work proposes a novel vendor allocation procedure to reduce the hardware cost that comes with Trojan detection methods. To further reduce the cost by minimizing resource requirements, an evolutionary algorithm-based Design Space Exploration methodology is proposed with options for loop unrolling and operation chaining. For reducing the cost of hardware Trojan detection and isolation, the proposed algorithm extends an existing implementation of Firefly algorithm. The proposed method is compared with the existing algorithms, using cost-based and Pareto-based evaluations. The results obtained demonstrate the ability of the new algorithm in achieving better solutions with a 77% reduction in cost when compared to the previous solutions.
29
Zhu, Shuping, Wei Gao et Xiaolei Li. « SSANet : normal-mode interference spectrum extraction via SSA algorithm-unrolled neural network ». Frontiers in Marine Science 10 (1 février 2024). http://dx.doi.org/10.3389/fmars.2023.1342090.
Texte intégralRésumé :
In ocean acoustic fields, extracting the normal-mode interference spectrum (NMIS) from the received sound intensity spectrum (SIS) plays an important role in waveguide-invariant estimation and underwater source ranging. However, the received SIS often has a low signal-to-noise ratio (SNR) owing to ocean ambient noise and the limitations of the received equipment. This can lead to significant performance degradation for the traditional methods of extracting NMIS at low SNR conditions. To address this issue, a new deep neural network model called SSANet is proposed to obtain NMIS based on unrolling the traditional singular spectrum analysis (SSA) algorithm. First, the steps of embedding and singular value decomposition (SVD) in SSA is achieved by the convolutional network. Second, the grouping step of the SSA is simulated using the matrix multiply weight layer, ReLU layer, point multiply weight layer and matrix multiply weight layer. Third, the diagonal averaging step was implemented using a fully connected network. Simulation results in canonical ocean waveguide environments demonstrate that SSANet outperforms other traditional methods such as Fourier transform (FT), multiple signal classification (MUSIC), and SSA in terms of root mean square error, mean absolute error, and extraction performance.
30
Peng, Guan‐Ju. « Learning the sparse prior : Modern approaches ». WIREs Computational Statistics 16, no 1 (janvier 2024). http://dx.doi.org/10.1002/wics.1646.
Texte intégralRésumé :
AbstractThe sparse prior has been widely adopted to establish data models for numerous applications. In this context, most of them are based on one of three foundational paradigms: the conventional sparse representation, the convolutional sparse representation, and the multi‐layer convolutional sparse representation. When the data morphology has been adequately addressed, a sparse representation can be obtained by solving the sparse coding problem specified by the data model. This article presents a comprehensive overview of these three models and their corresponding sparse coding problems and demonstrates that they can be solved using convex and non‐convex optimization approaches. When the data morphology is not known or cannot be analyzed, it must be learned from training data, thereby formulating dictionary learning problems. This article addresses two different dictionary learning paradigms. In an unsupervised scenario, dictionary learning involves the alternating or joint resolution of sparse coding and dictionary updating. Another option is to create a recurrent neural network by unrolling algorithms designed to solve sparse coding problems. These networks can then be used in a supervised learning setting to facilitate the training of dictionaries via forward‐backward optimization. This article lists numerous applications in various domains and outlines several directions for future research related to the sparse prior.This article is categorized under: Statistical Learning and Exploratory Methods of the Data Sciences > Modeling Methods Statistical and Graphical Methods of Data Analysis > Modeling Methods and Algorithms Statistical Models > Nonlinear Models
31
Heaton, Howard, Samy Wu Fung, Aviv Gibali et Wotao Yin. « Feasibility-based fixed point networks ». Fixed Point Theory and Algorithms for Sciences and Engineering 2021, no 1 (22 novembre 2021). http://dx.doi.org/10.1186/s13663-021-00706-3.
Texte intégralRésumé :
AbstractInverse problems consist of recovering a signal from a collection of noisy measurements. These problems can often be cast as feasibility problems; however, additional regularization is typically necessary to ensure accurate and stable recovery with respect to data perturbations. Hand-chosen analytic regularization can yield desirable theoretical guarantees, but such approaches have limited effectiveness recovering signals due to their inability to leverage large amounts of available data. To this end, this work fuses data-driven regularization and convex feasibility in a theoretically sound manner. This is accomplished using feasibility-based fixed point networks (F-FPNs). Each F-FPN defines a collection of nonexpansive operators, each of which is the composition of a projection-based operator and a data-driven regularization operator. Fixed point iteration is used to compute fixed points of these operators, and weights of the operators are tuned so that the fixed points closely represent available data. Numerical examples demonstrate performance increases by F-FPNs when compared to standard TV-based recovery methods for CT reconstruction and a comparable neural network based on algorithm unrolling. Codes are available on Github: github.com/howardheaton/feasibility_fixed_point_networks.
32
Alcibahy, Yasmine, Nicola G. Ghazi, Arif O. Khan et Aniruddha Agarwal. « Surgical repair of macular fold in X-linked retinoschisis initially misdiagnosed as familial exudative vitreoretinopathy ». RETINAL Cases & ; Brief Reports, 7 mai 2024. http://dx.doi.org/10.1097/icb.0000000000001600.
Texte intégralRésumé :
Purpose: To describe the presentation and surgical management of a young boy initially thought to have familial exudative vitreoretinopathy who was ultimately diagnosed with an unusually aggressive form of X-linked retinoschisis that included rapidly progressive bullous retinoschisis and tractional macular fold. Methods: Retrospective case report Results: A 19-month-old boy with straightening of major arcades, peripheral retinal ischemia, and in the left eye, a large macular fold was initially diagnosed as familial exudative vitreoretinopathy. During follow-up, he developed a rapidly progressive bullous retinoschisis in the left eye involving the inferior macula extending superiorly up to the macular fold. This revised the working diagnosis to X-linked retinoschisis, which was confirmed by genetic testing. Pars plana vitrectomy, inner flap retinectomy, unrolling of the macular fold and inner flap retinectomy, and C3F8 gas tamponade were performed. This resolved the macular fold. The patient showed good anatomical results without surgical complications up to 18 months of post-operative follow-up. Conclusions: X-linked retinoschisis can rarely present in young children with macular fold and peripheral ischemia, mimicking familial exudative vitreoretinopathy. Rapidly progressive bullous retinoschisis in this setting can be treated with pars plana vitrectomy, inner wall retinectomy, and removal of the vitreous traction to improve visual prognosis.
33
Chen, Hongling, Mauricio Sacchi, Hojjat Haghshenas Lari, Jinghuai Gao et Xiudi Jiang. « Nonstationary Seismic Reflectivity Inversion Based on Prior-engaged Semi-supervised Deep Learning Method ». GEOPHYSICS, 23 septembre 2022, 1–72. http://dx.doi.org/10.1190/geo2022-0057.1.
Texte intégralRésumé :
Reflectivity inversion methods based on a stationary convolution model are essential for seismic data processing. They compress the seismic wavelet, and by broadening the bandwidth of seismic data, they assist the interpretation of seismic sections. Unfortunately, they do not apply to realistic nonstationary deconvolution cases where the seismic wavelet varies as it propagates in the subsurface. Deep learning techniques have been proposed to solve inverse problems where networks can behave as the regularizer of the inverse problem. Our goal is to adopt a semi-supervised deep learning approach to invert reflectivity when the propagating wavelet is considered unknown and time-variant. To this end, we design a prior-engaged neural network by unrolling an alternating iterative optimization algorithm, where convolutional neural networks are used to solve two sub-problems. One is to invert the reflectivity, and the other is to estimate the time-varying wavelets. Generally, it is well-known that when working with geophysical inverse problems such as the one at hand, one has limited access to labeled data for training the network. We circumvent the problem by training the network via a data-consistency cost function where seismic traces are honored. Reflectivity estimates are also honored at spatial coordinates where true reflectivity series derived from borehole data are available. The cost function also penalizes time-varying wavelets from varying abruptly along the spatial direction. Experiments are conducted to show the effectiveness of the proposed method. We also compared the proposed approach to a nonstationary blind deconvolution algorithm based on regularized inversion. Our findings show that the proposed method improves the vertical resolution of seismic sections with noticeable correlation coefficient improvements over the nonstationary blind deconvolution. In addition, the proposed method is less sensitive to initial estimates of nonstationary wavelets. Moreover, it needs less human intervention when setting parameters than regularized inversion.
34
Waheed Muhammad SANYA, Gaurav BAJPAI et Haji Ali HAJI. « Implementation and Optimization of Image Processing on the Map of SABRE i.MX_6 ». International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 15 décembre 2021, 402–17. http://dx.doi.org/10.32628/cseit217690.
Texte intégralRésumé :
Vision relieves humans to understand the environmental deviations over a period. These deviations are seen by capturing the images. The digital image plays a dynamic role in everyday life. One of the processes of optimizing the details of an image whilst removing the random noise is image denoising. It is a well-explored research topic in the field of image processing. In the past, the progress made in image denoising has advanced from the improved modeling of digital images. Hence, the major challenges of the image process denoising algorithm is to advance the visual appearance whilst preserving the other details of the real image. Significant research today focuses on wavelet-based denoising methods. This research paper presents a new approach to understand the Sobel imaging process algorithm on the Linux platform and develop an effective algorithm by using different optimization techniques on SABRE i.MX_6. Our work concentrated more on the image process algorithm optimization. By using the OpenCV environment, this paper is intended to simulate a Salt and Pepper noisy phenomenon and remove the noisy pixels by using Median Filter Algorithm. The Sobel convolution method included and used in the design of a Sobel Filter and then process the image following the median filter, to achieve an effective edge detection result. Finally, this paper optimizes the algorithm on SABRE i.MX_6 Linux environment. By using algorithmic optimization (lower complexity algorithm in the mathematical sense, using appropriate data structures), optimization for RISC (loop unrolling) processors, including optimization for efficient use of hardware resources (access to data, cache management and multi-thread), this paper analyzed the different response parameters of the system with varied inputs, different compiler options (O1, O2, or O3), and different doping degrees. The proposed denoising algorithm shows the meaningful addition of the visual quality of the images and the algorithmic optimization assessment.