Abstract Surface defect detection is an essential task for ensuring the quality of products. Many excellent object detectors have been employed to detect surface defects in resent years, which has achieved outstanding success. To further improve performance, a detector based on state‐of‐the‐art YOLOv8, named improved YOLOv8 by neck, head and data (NHD‐YOLO), proposed. Specifically, from three crucial aspects including improved. First, shortcut feature pyramid network designed effectively...

Multi-objective scheduling problems in workshops are commonly encountered challenges the increasingly competitive market economy. These require a trade-off among multiple objectives such as time, energy consumption, and product quality. The importance of each optimization objective typically varies different time periods or contexts, necessitating decision-makers to devise optimal plans accordingly. In actual production, confront intricate multi-objective that demand balancing clients’...

Network intrusion detection systems are an important defense technology to guarantee information security and protect a network from attacks. In recent years, the broad learning system has attracted much attention been introduced into with some success. However, since traditional is simple linear structure, when dealing imbalanced datasets, it often ignores feature of minority class samples, leading poorer recognition rate samples. Secondly, high dimensionality redundant features in datasets...

The frequent occurrence of food safety accidents at the global level has triggered consumer sensitivity. Establishing a traceability system can effectively guarantee and increase confidence satisfaction. Existing systems often ignore environmental factors that affect quality all stages supply chain, authenticity information obtained through is difficult to guarantee. In this study, chain model was established based on blockchain radio frequency identification (RFID) technologies. focused...

Metallic surface defect detection is critical to ensure the quality of industrial products. Recently, human-advanced algorithms have been proposed. Most these rely on convolutional neural networks (CNN) and an anchoring scheme. However, a convolution unit only samples input feature maps at fixed shapes locations. Similarly, set anchors are uniformly predefined with scales shapes, which increases difficulties bounding box regression. Therefore, we propose adaptive anchor network for metallic...

Abstract To the problem of complex pre-processing and post-processing to obtain head-position existing in current crowd localization method using pseudo boundary box pre-designed positioning map, this work proposes an end-to-end framework named WSITrans, which reformulates weakly-supervised based on Transformer implements counting. Specifically, we first perform global maximum pooling (GMP) after each stage pure Transformer, can extract retain more detail heads. In addition, design a...

Recently, with the development of COVID-19 pandemic and next generation telecommunication network (5G), metaverse, in which users interact each other through avatars, has attracted increasingly attention. However, like traditional cyberspace, a three-dimensional (3D) virtual world built on internet, also faces various unexpected security threats. In addition, some forms methods cyberspace threats have undergone further evolution metaverse environment. Consequently, social engineering attacks...

As the distributed computing applications represent comprehensive dynamics and user requirements in them are significant difference, they require dynamic differentiated fine-grained resource access control policies taking context factors into consideration before making decisions. With regards to this, paper investigates impact functions of policy decision, proposes a context-aware method. Firstly, is analyzed classifiedly formalized from four aspects including platform security context,...

The frequent occurrence of food safety accidents and the globalization import export circulation make it very important to establish a traceability system. Due bulk low‐value characteristics grain, is difficult achieve in single unit like meat products; as grain has longer supply chain, involving much more links even complicated factors, easy cause information disconnection. In response above problems, this paper done following tasks: First, propose RFID‐based chain model, which mainly...

Spectrum-based Fault localization have proven to be useful in the process of software testing and debugging. However, how improve effectiveness fault has always been a research hot spot field engineering. Dynamic slicing can extract program dependencies under certain conditions. Thus, this technology is expected benefit for locating fault.We propose an improved dynamic spectrum-based general framework. We first obtain slice execution. Secondly, we construct mixed spectrum matrix from each...

Debugging programs involves a considerable amount of time and resources spent on identifying the source errors. Spectrum-based fault localization techniques are becoming increasingly popular due to their quick effective performance when dealing with that have single fault. However, these methods do not take into account reality most tend multiple faults. To address above challenges, we propose an automatic approach using ant colony algorithm via genetic operation for (ACOMFL). Our first...

We discuss the delay-constrained energy efficient multicast routing problem in cognitive radio ad hoc networks. A network is a wireless multihop established by secondary users on varying available spectrum bands. The research hot topic field of However, algorithm study networks still at its start stage. In this paper, we proposed novel (DCEEMR) DCEEMR guarantees that tree can be found if it exists, and satisfies delay bound has low consumption. uses delay-energy function to construct based...

Compared to traditional platform environments in the online realm, metaverse, as a three-dimensional (3D) virtual world, exposes more identity data network. Once these are compromised, it leads privacy breaches. Therefore, how ensure security metaverse environment has become an urgent problem be solved. Although research on authentication schemes can help improve security, network studied based their own environmental characteristics, which makes difficult meet needs environment. As result,...

The traditional software defect localisation method based on spectrum performs well in single scenarios, but it lacks effectiveness multiple scenarios. Cluster analysis is a to improve the accuracy of multi-defect locations. This clusters failed test cases generate centered around defects. By utilizing clustering results, reports are generated identify locations We proposed iterative self-organizing for multiple-fault (ISOMFL) efficacy localisation. have achieved this through some...