Abstract Convolutional Neural Networks (CNN) have immense potential to solve a broad range of computer vision problems. It has achieved encouraging results in numerous applications of engineering, medical, and other research fields due to the advancement in hardware, data collection procedures, and efficient algorithms. These innovations have changed the way how specific problems are solved as compared to conventional methods. This article presents a review of CNN implementation on civil structure crack detection. The review highlights the significant research that has been performed to detect structure cracks through classification and segmentation of crack images with CNN in the perspective of image pre-processing techniques, processing hardware, software tools, datasets, network architectures, learning procedures, loss functions, and network performance. The key contribution of this review article is the study and analysis of the most recent developments on crack detection using CNN. Additionally, this work also presents a discussion on crack detection through a manual process, image processing techniques, and machine learning methods along with their limitations. Finally, this article aims for assisting the readers to understand the motivation and methodology of the various CNN-based crack detection methods and to invoke them for exploring the solutions of challenges outlined in future research.
The concept of life skills is related to the way of life that emphasises the mutual exchange of knowledge, attitudes, and interpersonal skills in education. Its objective is to develop diverse skills among students and prepare them to face life’s challenges with determination. The World Health Organization has defined life skills as “the positive behaviours and tendencies that enable a person to adapt in day-to-day life.” Life skills are the abilities that enable a person to adapt and exhibit positive behaviour, allowing them to deal effectively with the problems and challenges of daily life. Life is a unique gift. Therefore, by equipping life with various skills, happiness, peace, and prosperity are created. In this research, with the objectives of the study in mind, an analytical examination of life skills among secondary-level students has been conducted. This research study examines the effects of living conditions, gender, and social class on students’ life skills and presents the findings. Future researchers can build upon this, and other factors affecting the research can also be explored.
Spiking Neural Networks (SNNs) are increasingly favored for deployment on resource-constrained edge devices due to their energy-efficient and event-driven processing capabilities. However, training SNNs remains challenging because of the computational intensity of traditional backpropagation algorithms adapted for spike-based systems. In this paper, we propose a novel software-hardware co-design that introduces a hardware-friendly training algorithm, Spiking Direct Feedback Alignment (SDFA) and implement it on a Resistive Random Access Memory (RRAM)-based In-Memory Computing (IMC) architecture, referred to as PipeSDFA, to accelerate SNN training. Software-wise, the computational complexity of SNN training is reduced by the SDFA through the elimination of sequential error propagation. Hardware-wise, a three-level pipelined dataflow is designed based on IMC architecture to parallelize the training process. Experimental results demonstrate that the PipeSDFA training accelerator incurs less than 2% accuracy loss on five datasets compared to baselines, while achieving 1.1X~10.5X and 1.37X~2.1X reductions in training time and energy consumption, respectively compared to PipeLayer.
External knowledge has played a crucial role in the recent development of computer use agents. We identify a critical knowledge-execution gap: retrieved knowledge often fails to translate into effective real-world task execution. Our analysis shows even 90% correct knowledge yields only 41% execution success rate. To bridge this gap, we propose UI-Evol, a plug-and-play module for autonomous GUI knowledge evolution. UI-Evol consists of two stages: a Retrace Stage that extracts faithful objective action sequences from actual agent-environment interactions, and a Critique Stage that refines existing knowledge by comparing these sequences against external references. We conduct comprehensive experiments on the OSWorld benchmark with the state-of-the-art Agent S2. Our results demonstrate that UI-Evol not only significantly boosts task performance but also addresses a previously overlooked issue of high behavioral standard deviation in computer use agents, leading to superior performance on computer use tasks and substantially improved agent reliability.
The graph parameter treedepth is minor-monotone; hence, the class of graphs with treedepth at most $k$ is minor-closed. By the Graph Minor Theorem, such a class is characterized by a finite set of forbidden minors. A conjecture of Dvořák, Giannopoulou, and Thilikos states that every such forbidden minor has at most $2^k$ vertices. We present an algorithm that, given $n, k \in \mathbb{N}$, computes the set of forbidden minors, forbidden subgraphs, and forbidden induced subgraphs on at most $n$ vertices, for the class of graphs of treedepth at most $k$. Applying this algorithm to $k = 4$ and $n = 16$, we enumerate 1546 forbidden minors, 1718 forbidden subgraphs, and 12204 forbidden induced subgraphs. Assuming the above conjecture holds, these sets constitute the complete obstruction sets for graphs of treedepth at most 4.
Sundaravadivazhagan Balasubaramanian, Robin Cyriac, Sahana Roshan
et al.
The COVID-19 pandemic has been infecting the entire world over the past years. To prevent the spread of COVID-19, people have acclimatised to the new normal, which includes working from home, communicating online, and maintaining personal cleanliness. There are numerous tools required to prepare to compact transmissions in the future. One of these elements for protecting individuals from fatal virus transmission is the mask. Studies have indicated that wearing a mask may help to reduce the risk of viral transmission of all kinds. It causes many public places to take efforts to ensure that its guests wear adequate face masks and keep a safe distance from one another. Screening systems need to be installed at the doors of businesses, schools, government buildings, private offices, and/or other important areas. A variety of face detection models have been designed using various algorithms and techniques. Most of the articles in the previously published research have not worked on dimensionality reduction in conjunction with depth-wise separable neural networks. The necessity of determining the identities of people who do not cover their faces when they are in public is the driving factor for the development of this methodology. This research work proposes a deep learning technique to determine if a person is wearing mask or not and identifies whether it is properly worn or not. Stacked Auto Encoder (SAE) technique is implemented by stacking the following components: Principal Component Analysis (PCA) and Depth-wise Separable Convolutional Neural Network (DWSC-NN). PCA is used to reduce the irrelevant features in the images and resulted high true positive rate in the detection of mask. We achieved an accuracy score of 94.16% and an F1 score of 96.009% by the application of the method described in this research.
Zsuzsanna Pálffy, Lívia Ablonczy-Mihályka, Petra Kecskés
Corporate embeddedness and sustainable corporate behaviour are closely linked. Consequently, recent literature has introduced the term sustainability embeddedness. However, the concept is still at an early stage, and both its conceptualisation and practical implications remain incomplete. From the literature review, it can be determined that the term sustainability embeddedness can be classified as one of the so-called fuzzy concepts because its use in different literatures is associated with different meanings. The main goal of the study is to review the previous interpretations and to contribute to the literature on the concept with a specific explanation, placing it in the set of previous definitions. A further aim of the research is to identify the best sustainable practices of corporate embeddedness using the quintuple helix model. The goal is to demonstrate how sustainability practices can enhance mutually beneficial relationships between companies and other actors within the local space. The study serves as a literature basis for later research, in which, as a continuation of the collection of practices, small and medium-sized family businesses in Gyor, Hungary, will be examined. Along the theoretical model of the study, practical implications can be identified as they may serve as an incentive practice for regional small businesses, which helps to deepen their degree of embeddedness, thus exploiting additional local benefits for the local space.
Chemical engineering, Computer engineering. Computer hardware
The era of big data network represented mainly by the Internet of Things is limited by various factors such as various environments, volumes, and calculations. This paper proposes and studies a human motion detection system based on the intelligent computing of the Internet of Things, which can effectively detect the daily motion of the human body. In the era of the Internet of Things, this paper designs and develops a human motion detection system based on the Internet of Things technology and intelligent computing and explores the law of normal human motion. This paper also analyzes the design ideas and technical advantages of the human motion detection system from many aspects. This research is a human motion detection system designed and developed under the comprehensive use of Internet of Things technology and intelligent computing technology, which can effectively detect the actual situation of the human body in the state of motion. Experimental results show that the overall accuracy of the system for monitoring and recognizing human walking is 96.91%, and the overall accuracy of monitoring and recognizing human jogging is 97.18%. It monitors and recognizes human jogging with an overall accuracy rate of 97.96%. It has great practical significance.
Nirmal Choudhary, Henning Clever, Robert Ludwigs
et al.
The increasing global demand for high‐quality and low‐cost battery electrodes poses major challenges for battery cell production. As mechanical defects on the electrode sheets have an impact on the cell performance and their lifetime, inline quality control during electrode production is of high importance. Correlation of detected defects with process parameters provides the basis for optimization of the production process and thus enables long‐term reduction of reject rates, shortening of the production ramp‐up phase, and maximization of equipment availability. To enable automatic detection of visually detectable defects on electrode sheets passing through the process steps at a speed of 9 m s−1, a You‐Only‐Look‐Once architecture (YOLO architecture) for the identification of visual detectable defects on coated electrode sheets is demonstrated within this work. The ability of the quality assurance (QA) system developed herein to detect mechanical defects in real time is validated by an exemplary integration of the architecture into the electrode manufacturing process chain at the Battery Lab Factory Braunschweig.
Computer engineering. Computer hardware, Control engineering systems. Automatic machinery (General)
Yanis Belkheyar, Joan Daemen, Christoph Dobraunig
et al.
Recently, a memory safety concept called Cryptographic Capability Computing (C3) has been proposed. C3 is the first memory safety mechanism that works without requiring extra storage for metadata and hence, has the potential to significantly enhance the security of modern IT-systems at a rather low cost. To achieve this, C3 heavily relies on ultra-low-latency cryptographic primitives. However, the most crucial primitive required by C3 demands uncommon dimensions. To partially encrypt 64-bit pointers, a 24-bit tweakable block cipher with a 40-bit tweak is needed. The research on low-latency tweakable block ciphers with such small dimensions is not very mature. Therefore, designing such a cipher provides a great research challenge, which we take on with this paper. As a result, we present BipBip, a 24-bit tweakable block cipher with a 40-bit tweak that allows for ASIC implementations with a latency of 3 cycles at a 4.5 GHz clock frequency on a modern 10 nm CMOS technology.
Computer engineering. Computer hardware, Information technology
Location tracking with global navigation satellite systems (GNSS), such as the GPS, is used in many applications, including the tracking of wild animals for research. Snapshot GNSS is a technique that only requires milliseconds of satellite signals to infer the position of a receiver. This is ideal for low-power applications such as animal tracking. However, there are few existing snapshot systems, none of which is open source. To address this, we developed SnapperGPS, a fully open-source, low-cost, and low-power location tracking system designed for wildlife tracking. SnapperGPS comprises three parts, all of which are open-source: (i) a small, low-cost, and low-power receiver; (ii) a web application to configure the receiver via USB; and (iii) a cloud-based platform for processing recorded data. This paper presents the hardware side of this project. The total component cost of the receiver is under $30, making it feasible for field work with restricted budgets and low recovery rates. The receiver records very short and low-resolution samples resulting in particularly low power consumption, outperforming existing systems. It can run for more than a year on a 40 mAh battery. We evaluated SnapperGPS in controlled static and dynamic tests in a semi-urban environment where it achieved median errors of 12 m. Additionally, SnapperGPS has already been deployed for two wildlife tracking studies on sea turtles and sea birds.
In this paper, we propose TensorFHE, an FHE acceleration solution based on GPGPU for real applications on encrypted data. TensorFHE utilizes Tensor Core Units (TCUs) to boost the computation of Number Theoretic Transform (NTT), which is the part of FHE with highest time-cost. Moreover, TensorFHE focuses on performing as many FHE operations as possible in a certain time period rather than reducing the latency of one operation. Based on such an idea, TensorFHE introduces operation-level batching to fully utilize the data parallelism in GPGPU. We experimentally prove that it is possible to achieve comparable performance with GPGPU as with state-of-the-art ASIC accelerators. TensorFHE performs 913 KOPS and 88 KOPS for NTT and HMULT (key FHE kernels) within NVIDIA A100 GPGPU, which is 2.61x faster than state-of-the-art FHE implementation on GPGPU; Moreover, TensorFHE provides comparable performance to the ASIC FHE accelerators, which makes it even 2.9x faster than the F1+ with a specific workload. Such a pure software acceleration based on commercial hardware with high performance can open up usage of state-of-the-art FHE algorithms for a broad set of applications in real systems.
Approximate computing (AxC) has been long accepted as a design alternative for efficient system implementation at the cost of relaxed accuracy requirements. Despite the AxC research activities in various application domains, AxC thrived the past decade when it was applied in Machine Learning (ML). The by definition approximate notion of ML models but also the increased computational overheads associated with ML applications-that were effectively mitigated by corresponding approximations-led to a perfect matching and a fruitful synergy. AxC for AI/ML has transcended beyond academic prototypes. In this work, we enlighten the synergistic nature of AxC and ML and elucidate the impact of AxC in designing efficient ML systems. To that end, we present an overview and taxonomy of AxC for ML and use two descriptive application scenarios to demonstrate how AxC boosts the efficiency of ML systems.
T. Schlick, Stephanie Portillo-Ledesma, Mischa Blaszczyk
et al.
Modeling and simulation have quickly become equivalent pillars of research along with traditional theory and experimentation. The growing realization that most complex phenomena of interest span many orders of spatial and temporal scales has led to an exponential rise in the development and application of multiscale modeling and simulation over the past two decades. In this perspective, the associate editors of the International Journal for Multiscale Computational Engineering and their co-workers illustrate current applications in their respective fields spanning biomolecular structure and dynamics, civil engineering and materials science, computational mechanics, aerospace and mechanical engineering, and more. Such applications are highly tailored, exploit the latest and ever-evolving advances in both computer hardware and software, and contribute significantly to science, technology, and medical challenges in the 21st century.
Naweiluo Zhou, Yiannis Georgiou, Marcin Pospieszny
et al.
Abstract Containerisation demonstrates its efficiency in application deployment in Cloud Computing. Containers can encapsulate complex programs with their dependencies in isolated environments making applications more portable, hence are being adopted in High Performance Computing (HPC) clusters. Singularity, initially designed for HPC systems, has become their de facto standard container runtime. Nevertheless, conventional HPC workload managers lack micro-service support and deeply-integrated container management, as opposed to container orchestrators. We introduce a Torque-Operator which serves as a bridge between HPC workload manager (TORQUE) and container orchestrator (Kubernetes). We propose a hybrid architecture that integrates HPC and Cloud clusters seamlessly with little interference to HPC systems where container orchestration is performed on two levels.
To effectively use the user-item interaction history and auxiliary information in recommendation systems,this paper proposes an improved collaborative filtering recommendation algorithm.Based on semi-autoencoder,the features of auxiliary information of users and items are extracted,and then mapped into the Matrix Factorization(MF) model.By using the back propagation algorithm,the semi-autoencoder and the matrix factorization model are jointly updated to improve the recommendation performance.Experimental results on the public datasets of MovieLens-100K and Book-Crossing show that the proposed algorithm provides better recommendation effects than the traditional recommendation algorithms,including the Biased Singular Value Decomposition(Biased SVD) and the Probabilistic Matrix Factorization(PMF) algorithm.