The combination of chaos theory and cryptographic methodologies constitutes a pivotal aspect of the information security. Owing to certain attributes of images, like their large data capacity and significant redundancy, image encryption needs specialized techniques instead of conventional text encryption. This study presents an innovative hybrid image encryption scheme that integrates Henon map–based key generation and involves diffusion process to ensure security and efficiency of the system. The experimental findings indicate that the methodology provides strong confusion and diffusion, low correlation between adjacent pixels, high entropy values close to 8, and number of pixel change rate (NPCR) above 99%. The method ensures secure image transmission using the same secret keys and maintains computational efficiency, making it suitable for practical applications in secure communication and multimedia protection. Unlike conventional Henon map–based encryption schemes, this study fuses Henon map–based key generation with error correction code (ECC) and an organized diffusion mechanism, enhancing both cryptographic robustness and reliability against channel errors during transmission.
Based on the mechanism of human visual imaging, retinal projection display (RPD) directly projects images onto the retina, achieving sharp images without relying on the focal adjustment of human eye. However, the physiological phenomenon of eye movements makes it difficult to align the convergence point of image lights with the human pupil, especially when the viewer needs to wear a pair of vision‐corrected frame glasses, resulting in blurred images. To accommodate the movements of human eye, herein, a waveguide‐based RPD system with bidirectional extended eyebox is proposed, in which two‐layer holographic optical elements (HOEs) are designed as the image combiner to generate different eye reliefs along the visual axis of the human eye. Each layer of HOEs generates two horizontally arranged viewpoints, thus, achieving bidirectional eyebox expansion. Experimental results show that the proposed RPD system provides two sets of viewpoints with the eye reliefs of 11 and 12 mm, respectively, and obtains a 2 × 2 viewpoint array distributed horizontally and axially. Additionally, two sets of viewpoints can be switched to accommodate the different eyewear habits of viewers. The proposed RPD system enhances the adaptability of near‐eye display device to the human eye through the bidirectional eyebox expansion.
Computer engineering. Computer hardware, Control engineering systems. Automatic machinery (General)
W. Michael Brown, Anurag Ramesh, Thomas Lubinski
et al.
As is intrinsic to the fundamental goal of quantum computing, classical simulation of quantum algorithms is notoriously demanding in resource requirements. Nonetheless, simulation is critical to the success of the field and a requirement for algorithm development and validation, as well as hardware design. GPU-acceleration has become standard practice for simulation, and due to the exponential scaling inherent in classical methods, multi-GPU simulation can be required to achieve representative system sizes. In this case, inter-GPU communications can bottleneck performance. In this work, we present the introduction of MPI into the QED-C Application-Oriented Benchmarks to facilitate benchmarking on HPC systems. We review the advances in interconnect technology and the APIs for multi-GPU communication. We benchmark using a variety of interconnect paths, including the recent NVIDIA Grace Blackwell NVL72 architecture that represents the first product to expand high-bandwidth GPU-specialized interconnects across multiple nodes. We show that while improvements to GPU architecture have led to speedups of over 4.5X across the last few generations of GPUs, advances in interconnect performance have had a larger impact with over 16X performance improvements in time to solution for multi-GPU simulations.
Conventional heterogeneous computing systems built on PCIe interconnects suffer from inefficient fine-grained host-device interactions and complex programming models. In recent years, many proprietary and open cache-coherent interconnect standards have emerged, among which compute express link (CXL) prevails in the open-standard domain after acquiring several competing solutions. Although CXL-based coherent heterogeneous computing holds the potential to fundamentally transform the collaborative computing mode of CPUs and XPUs, research in this direction remains hampered by the scarcity of available CXL-supported platforms, immature software/hardware ecosystems, and unclear application prospects. This paper presents Cohet, the first CXL-driven coherent heterogeneous computing framework. Cohet decouples the compute and memory resources to form unbiased CPU and XPU pools which share a single unified and coherent memory pool. It exposes a standard malloc/mmap interface to both CPU and XPU compute threads, leaving the OS dealing with smart memory allocation and management of heterogeneous resources. To facilitate Cohet research, we also present a full-system cycle-level simulator named SimCXL, which is capable of modeling all CXL sub-protocols and device types. SimCXL has been rigorously calibrated against a real CXL testbed with various CXL memory and accelerators, showing an average simulation error of 3%. Our evaluation reveals that CXL.cache reduces latency by 68% and increases bandwidth by 14.4x compared to DMA transfers at cacheline granularity. Building upon these insights, we demonstrate the benefits of Cohet with two killer apps, which are remote atomic operation (RAO) and remote procedure call (RPC). Compared to PCIe-NIC design, CXL-NIC achieves a 5.5 to 40.2x speedup for RAO offloading and an average speedup of 1.86x for RPC (de)serialization offloading.
Due to their powerful sequence modeling capabilities, Transformer-based image captioning models have demonstrated remarkable performance. However, most of these models typically utilize region visual features to perform encoding and decoding, which cannot fully use the fine-grained information of the whole image, and this leads to visual feature confusion. Accordingly, we introduce panoptic segmentation into the Transformer-based image captioning model by replacing the region visual feature with mask visual features and propose a novel image captioning model based on multi-visual-feature fusion. Our model not only disentangles the region visual features effectively but also makes use of both mask and grid visual features to improve image captioning performance. We perform quantitative and qualitative experiments on the MSCOCO dataset, which demonstrate that our method significantly outperforms existing Transformer-based image captioning models. In addition, our model enhances the interpretability of the caption generation process, and more specifically, achieves CIDEr and BLEU-4 scores of 138.5 and 41, respectively.
Nguyet T. M. Nguyen, Huyen K. M. Tran, Trang K. M. Tran
et al.
Wheat Germ (WG) is a by-product that is rich in nutritional value but is facing the problem of food loss and waste. This study utilised WG from the wheat mill to process vegetarian sausage according to traditional processing methods to investigate the effect of added WG ratio on the quality of vegetarian sausage, including structural properties, sensory value and nutritional value at different mixing ratios. Additional WG rates were examined at 0, 4, 5, 6, 7, and 8 % (% wt). Evaluate the textural properties using the TPA method on the Textural analyzer, SEM microstructure, retained water, pH value, and colour parameter combined with sensory value analysis. The results showed that WG can improve hardness, adhesiveness, gumminess and chewiness, but cohesiveness and springiness were not significant. The pH index gradually decreased, the water separation ability changed clearly, and the colour values also changed, especially the a* and b* values, which decreased when adding WG. Although there are still some low outliers, the sensory scores of the 6 and 7 % added vegetarian sausage have an average score of about 5.5, comparable to the control sample. Besides, WG has also been proven to improve nutritional value without adding artificial sweeteners, especially protein composition. This work can be considered for utilizing WG to develop products rich in plant protein.
Chemical engineering, Computer engineering. Computer hardware
Background It is currently observed that some landfills are experiencing severe overloading, with some having ceased operations. However, they continue to threaten the environment and public health. There is an urgent need for governance, although the process is complex and requires more intelligent and efficient governance approaches. Methods This study explored the application of digital twin technology based on multi-agent systems in the environmental governance of abandoned landfills. This paper addresses the demands of landfill governance by integrating modules, including twin models, mechanisms, and big data, and integrating each module with corresponding intelligent agents, forming a thoughtful, collaborative, and adaptive digital twin agent system. Results This method can collect and analyze on-site data more systematically and provide feedback to management personnel to guide the adjustment of on-site plans and improve the on-site management efficiency by 30%. Conclusions Through application cases, the operation process of this system in specific landfill environmental governance scenarios was demonstrated, confirming its superiority in environmental governance. This system can facilitate environmental monitoring, intelligent analysis, and decision control during the governance of abandoned landfills.
A limitation of brain-computer interface (BCI) spellers is that they require the user to be able to move the eyes to fixate on targets. This poses an issue for users who cannot voluntarily control their eye movements, for instance, people living with late-stage amyotrophic lateral sclerosis (ALS). This pilot study makes the first step towards a gaze-independent speller based on the code-modulated visual evoked potential (c-VEP). Participants were presented with two bi-laterally located stimuli, one of which was flashing, and were tasked to attend to one of these stimuli either by directly looking at the stimuli (overt condition) or by using spatial attention, eliminating the need for eye movement (covert condition). The attended stimuli were decoded from electroencephalography (EEG) and classification accuracies of 88% and 100% were obtained for the covert and overt conditions, respectively. These fundamental insights show the promising feasibility of utilizing the c-VEP protocol for gaze-independent BCIs that use covert spatial attention when both stimuli flash simultaneously.
Abstract Video has become an essential medium to monitoring, identification and knowledge sharing. For industrial applications, especially Industrial Internet of Things (IIoT), videos encoded with specific protocols are transferred to smart gateways. In a typical IIoT scenario, the protocol of the video is firstly recognised, which prepares for subsequent video tasks. Due to the constrained resources in such scenarios, the video quality can be deteriorated during encoding and compression processes, which is challenging for IIoT. Recently, there have been extensive works focussing on the protocol identification (PI) and video quality enhancement (VQE) tasks on IIoT edge devices using deep neural networks (DNNs). Since DNNs often require high computational resources, complex networks can hardly be deployed on edge devices. An IIoT system which can efficiently identify the stream protocol and enhance the video quality is proposed in this study. The light‐weighted network designs and inference optimisation techniques have been proposed for PI and VQE to realise efficient deployments. Our proposed system employed on an IIoT edge device can achieve an accuracy of higher than 97.52% with fast inference speed for PI. For the VQE task, our system has demonstrated superior performance (15.230 FPS, 0.773 FPS/W) in comparison with the state‐of‐the‐art methods.
In diesem Beitrag beschreiben wir ein Rollenspiel, das im Unterricht bereits ab der Sekundarstufe 1 mit der gesamten Klasse durchgeführt werden kann und das Prinzip des verstärkenden Lernens im Themenbereich „maschinelles Lernen“ erfahrbar macht.
Yongsheng QI, Xiaoxu DU, Junfeng ZHU, Shengli GAO, Liqiang LIU
Realizing big data to manage livestock requires real-time monitoring of livestock, but real-time monitoring of livestock is easily interfered by large changes in target size, lighting, environmental factors, etc., so it is difficult to detection, and existing livestock detection algorithms have the problem of poor robustness.An object detection network called E-YOLOv4-tiny is proposed based on enhanced YOLOv4-tiny, which adopts a pyramid network with multi-scale feature fusion, taking into account shallow local detail features and deep semantic information to solve the problem of livestock size fluctuation in pastoral areas.The number of backbone network parameters is reduced by improving the residual structure to accommodate embedded platform requirements.A new composite clustering algorithm is introduced to design anchor frames to improve the accuracy of the algorithm under the premise of ensuring portability.Finally, according to the characteristics of a pastoral environment, a new Compound Muti-channel Attention(CMA) mechanism is proposed to improve the poor accuracy of the target detection network and enhance the robustness of the algorithm.Experimental results show that the mean Average Precision(mAP) of the E-YOLOv4-tiny algorithm is 0.878 9, and the frame rate is 32 frame/s, and it's mAP is 9.32% higher than that of the traditional YOLOv4-tiny algorithm while maintaining almost the same detection rate.
Image filtering algorithms have wide applications in such fields as machine learning, image processing, and image recognition.They play an important role in reducing "salt and pepper" noise, image binarization, edge recognition, and feature extraction. Although common image filtering algorithms are implemented in the OpenCV open source library, a significant gap in performance exists compared with other platforms on the Android platform. With the rapid development of embedded platforms, the performance requirements for filtering algorithms on embedded platforms have become increasingly high in practical applications. Therefore, starting with filtering algorithms with wide application scenarios, such as morphological filtering, box filtering, threshold filtering, compression filtering, and arithmetic filtering, a series of high-performance image filtering algorithms designed for the Android platform based on OpenGL ES are developed and implemented. OpenGL ES calculation shaders are used to accelerate the algorithm in parallel, using texture objects for memory optimization, and in-depth optimization in image boundary processing, image data types, and data communication is conducted. This approach resulted in better performance. The optimized image filtering algorithm is compared with the corresponding algorithm in the open-source OpenCV library. The experimental results show that the overall performance of the image filtering algorithm based on the Android platform using the OpenGL ES interface is significantly better than the performances of the relevant algorithms in the OpenCV library. The larger the image size, the more obvious the computational advantage. The maximum performance improvement is 110.018 times that of the corresponding algorithm in the OpenCV library.
Researchers have long tried to minimize training costs in deep learning while maintaining strong generalization across diverse datasets. Emerging research on dataset distillation aims to reduce training costs by creating a small synthetic set that contains the information of a larger real dataset and ultimately achieves test accuracy equivalent to a model trained on the whole dataset. Unfortunately, the synthetic data generated by previous methods are not guaranteed to distribute and discriminate as well as the original training data, and they incur significant computational costs. Despite promising results, there still exists a significant performance gap between models trained on condensed synthetic sets and those trained on the whole dataset. In this paper, we address these challenges using efficient Dataset Distillation with Attention Matching (DataDAM), achieving state-of-the-art performance while reducing training costs. Specifically, we learn synthetic images by matching the spatial attention maps of real and synthetic data generated by different layers within a family of randomly initialized neural networks. Our method outperforms the prior methods on several datasets, including CIFAR10/100, TinyImageNet, ImageNet-1K, and subsets of ImageNet-1K across most of the settings, and achieves improvements of up to 6.5% and 4.1% on CIFAR100 and ImageNet-1K, respectively. We also show that our high-quality distilled images have practical benefits for downstream applications, such as continual learning and neural architecture search.
Chiara Caiazza, Claudio Cicconetti, Valerio Luconi
et al.
Edge computing is projected to become the dominant form of cloud computing in the future because of the significant advantages it brings to both users (less latency, higher throughput) and telecom operators (less Internet traffic, more local management). However, to fully unlock its potential at scale, system designers and automated optimization systems alike will have to monitor closely the dynamics of both processing and communication facilities. Especially the latter is often neglected in current systems since network performance in cloud computing plays only a minor role. In this paper, we propose the architecture of MECPerf, which is a solution to collect network measurements in a live edge computing domain, to be collected for offline provisioning analysis and simulations, or to be provided in real-time for on-line system optimization. MECPerf has been validated in a realistic testbed funded by the European Commission (Fed4Fire+), and we describe here a summary of the results, which are fully available as open data and through a Python library to expedite their utilization. This is demonstrated via a use case involving the optimization of a system parameter for migrating clients in a federated edge computing system adopting the GSMA platform operator concept.
Greenhouse gas emissions reduction in the industrial sector focusses on energy-intensive industries (EIIs) since they comprise a significant proportion of industrial sector emissions. However, collective emissions from non-EII industries is substantial and organisations require guidance to reach carbon neutrality. This study addresses the research question ‘What methodologies and modelling tools do industrial organisations use to plan and subsequently achieve carbon emissions neutrality?’ There is a research gap between general, independent, emissions abatement measures and organisation-specific plans developed using in-house or commercial software and/or energy management consultants. This study proposes a detailed, open access, cross-sectoral and strategic methodology aimed at the organisational level. A Supplier-Input-Process-Output-Customer (SIPOC) methodology is used in an adapted Define-Measure-Analyse-Improve-Control (DMAIC) framework to provide a high-level, visual pathway to carbon neutrality, clearly indicating the supplier, input, process, output, and customer for each carbon mitigation step. The Analyse step involves an energy audit, and heat and renewable energy studies to generate modelling tool input. The Improve step models the potential emissions abatement measures in priority order of efficiency, new technology, heat recovery, and renewables, with the budget or timeline as the dominant parameter. The model outputs are a carbon neutrality waterfall as the pathway, and a sensitivity graph to highlight influential modelling inputs. Future work includes modelling tool development, and validation with a case study of a medical device manufacturing facility.
Chemical engineering, Computer engineering. Computer hardware
Spiking Neural Networks (SNNs) compute in an event-based matter to achieve a more efficient computation than standard Neural Networks. In SNNs, neuronal outputs (i.e. activations) are not encoded with real-valued activations but with sequences of binary spikes. The motivation of using SNNs over conventional neural networks is rooted in the special computational aspects of SNNs, especially the very high degree of sparsity of neural output activations. Well established architectures for conventional Convolutional Neural Networks (CNNs) feature large spatial arrays of Processing Elements (PEs) that remain highly underutilized in the face of activation sparsity. We propose a novel architecture that is optimized for the processing of Convolutional SNNs (CSNNs) that feature a high degree of activation sparsity. In our architecture, the main strategy is to use less but highly utilized PEs. The PE array used to perform the convolution is only as large as the kernel size, allowing all PEs to be active as long as there are spikes to process. This constant flow of spikes is ensured by compressing the feature maps (i.e. the activations) into queues that can then be processed spike by spike. This compression is performed in run-time using dedicated circuitry, leading to a self-timed scheduling. This allows the processing time to scale directly with the number of spikes. A novel memory organization scheme called memory interlacing is used to efficiently store and retrieve the membrane potentials of the individual neurons using multiple small parallel on-chip RAMs. Each RAM is hardwired to its PE, reducing switching circuitry and allowing RAMs to be located in close proximity to the respective PE. We implemented the proposed architecture on an FPGA and achieved a significant speedup compared to other implementations while needing less hardware resources and maintaining a lower energy consumption.
Sanjib Ghosh, Tanjung Krisnanda, T. Paterek
et al.
Quantum computers require precise control over parameters and careful engineering of the underlying physical system. In contrast, neural networks have evolved to tolerate imprecision and inhomogeneity. Here, using a reservoir computing architecture we show how a random network of quantum nodes can be used as a robust hardware for quantum computing. Our network architecture induces quantum operations by optimising only a single layer of quantum nodes, a key advantage over the traditional neural networks where many layers of neurons have to be optimised. We demonstrate how a single network can induce different quantum gates, including a universal gate set. Moreover, in the few-qubit regime, we show that sequences of multiple quantum gates in quantum circuits can be compressed with a single operation, potentially reducing the operation time and complexity. As the key resource is a random network of nodes, with no specific topology or structure, this architecture is a hardware friendly alternative paradigm for quantum computation. Building quantum computers typically requires substantial engineering efforts to achieve precise control on qubits and quantum gates. Here, the authors introduce an architecture based on reservoir computing and machine learning to realize efficient quantum operations without resorting to full optimization of the control parameters.