The anchor-free object detector CenterNet regards the object as a center point and predicts it based on the Gaussian circle region. For each object's center point, CenterNet directly regresses the width and height of the objects and finally gets the boundary range of the objects. However, the critical range of the object's center point can not be accurately limited by using the Gaussian circle region to constrain the prediction region, resulting in many low-quality centers' predicted values. In addition, because of the large difference between the width and height of different objects, directly regressing the width and height will make the model difficult to converge and lose the intrinsic relationship between them, thereby reducing the stability and consistency of accuracy. For these problems, we proposed a center point prediction method based on the Gaussian elliptic region and a size component regression method based on the small solution space. First, we constructed a Gaussian ellipse region that can accurately predict the object's center point. Second, we recode the width and height of the objects, which significantly reduces the regression solution space and improves the convergence speed of the model. Finally, we jointly decode the predicted components, enhancing the internal relationship between the size components and improving the accuracy consistency. Experiments show that when using CenterNet as the improved baseline and Hourglass-104 as the backbone, on the MS COCO dataset, our improved model achieved 44.7%, which is 2.6% higher than the baseline.

High-definition (HD) maps can provide precise road information that enables an autonomous driving system to effectively navigate a vehicle. Recent research has focused on leveraging semantic segmentation to achieve automatic annotation of HD maps. However, the existing methods suffer from low recognition accuracy in automatic driving scenarios, leading to inefficient annotation processes. In this paper, we propose a novel semantic segmentation method for automatic HD map annotation. Our approach introduces a new encoder, known as the convolutional transformer hybrid encoder, to enhance the model's feature extraction capabilities. Additionally, we propose a multi-level fusion module that enables the model to aggregate different levels of detail and semantic information. Furthermore, we present a novel decoupled boundary joint decoder to improve the model's ability to handle the boundary between categories. To evaluate our method, we conducted experiments using the Bird's Eye View point cloud images dataset and Cityscapes dataset. Comparative analysis against stateof-the-art methods demonstrates that our model achieves the highest performance. Specifically, our model achieves an mIoU of 56.26%, surpassing the results of SegFormer with an mIoU of 1.47%. This innovative promises to significantly enhance the efficiency of HD map automatic annotation.

Accurate prediction of critical illness is significant for ensuring the lives and health of patients. The selection of indicators affects the real-time capability and accuracy of the prediction for critical illness. However, the diversity and complexity of these indicators make it difficult to find potential connections between them and critical illnesses. For the first time, this study proposes an indicator analysis model to extract key indicators from the preoperative and intraoperative clinical indicators and laboratory results of critical illnesses. In this study, preoperative and intraoperative data of heart failure and respiratory failure are used to verify the model. The proposed model processes the datum and extracts key indicators through four parts. To test the effectiveness of the proposed model, the key indicators are used to predict the two critical illnesses. The classifiers used in the prediction are light gradient boosting machine (LightGBM) and eXtreme Gradient Boosting (XGBoost). The predictive performance using key indicators is better than that using all indicators. In the prediction of heart failure, LightGBM and XGBoost have sensitivities of 0.889 and 0.892, and specificities of 0.939 and 0.937, respectively. For respiratory failure, LightGBM and XGBoost have sensitivities of 0.709 and 0.689, and specificity of 0.936 and 0.940, respectively. The proposed model can effectively analyze the correlation between indicators and postoperative critical illness. The analytical results make it possible to find the key indicators for postoperative critical illnesses. This model is meaningful to assist doctors in extracting key indicators in time and improving the reliability and efficiency of prediction.

Analysis of a bipartite (two-mode) network is a significant research area to understand the formation of social communities, economic systems, drug side effect topology, etc. in complex information systems. Most of the previous works talk about a projection-based model or latent feature model, which predicts the link based on singular similarity. The projection-based models suffer from the loss of structural information in the projected network and the latent feature is hardly present. This work proposes a novel method for link prediction in the bipartite network based on an ensemble of composite similarities, overcoming the issues of model-based and latent feature models. The proposed method analyzes the structure, neighborhood nodes as well as latent attributes between the nodes to predict the link in the network. To illustrate the proposed method, experiments are performed with five real-world data sets and compared with various state-of-art link prediction methods and it is inferred that this method outperforms with ~3% to ~9% higher using area under the precision-recall curve (AUC-PR) measure. This work holds great significance in the study of biological networks, e-commerce networks, complex web-based systems, networks of drug binding, enzyme protein, and other related networks in understanding the formation of such complex networks. Further, this study helps in link prediction and its usability for different purposes ranging from building intelligent systems to providing services in big data and web-based systems.

This paper proposes two video quality assessment methods based on deep neural network. (i)The first method uses the IQF-CNN (convolution neural network based on image quality features) to build image quality assessment method. The LIVE image database is used to test this method, the experiment show that it is effective. Therefore, this method is extended to the video quality assessment. At first every image frame of video is predicted, next the relationship between different image frames are analyzed by the hysteresis function and different window function to improve the accuracy of video quality assessment. (ii)The second method proposes a video quality assessment method based on convolution neural network (CNN) and gated circular unit network (GRU). First, the spatial features of video frames are extracted using CNN network, next the temporal features of the video frame using GRU network. Finally the extracted temporal and spatial features are analyzed by full connection layer of CNN network to obtain the video quality assessment score. All the above proposed methods are verified on the video databases, and compared with other methods.

With the rapid development of deep learning, Depth Map Super-Resolution (DMSR) method has achieved more advanced performances. However, when the upsampling rate is very large, it is difficult to capture the structural consistency between color features and depth features by these DMSR methods. Therefore, we propose a color-image guided DMSR method based on iterative depth feature enhancement. Considering the feature difference between high-quality color features and low-quality depth features, we propose to decompose the depth features into High-Frequency (HF) and Low-Frequency (LF) components. Due to structural homogeneity of depth HF components and HF color features, only HF color features are used to enhance the depth HF features without using the LF color features. Before the HF and LF depth feature decomposition, the LF component of the previous depth decomposition and the updated HF component are combined together. After decomposing and reorganizing recursively-updated features, we combine all the depth LF features with the final updated depth HF features to obtain the enhanced-depth features. Next, the enhanced-depth features are input into the multistage depth map fusion reconstruction block, in which the cross enhancement module is introduced into the reconstruction block to fully mine the spatial correlation of depth map by interleaving various features between different convolution groups. Experimental results can show that the two objective assessments of root mean square error and mean absolute deviation of the proposed method are superior to those of many latest DMSR methods.

The rise of Industry 5.0 has led to a smarter and more digital way of doing business, but with it comes the issue of user privacy and security. Only when privacy and security issues are addressed, will users be able to transact online with greater peace of mind. Thus, to address the security and privacy problems associated with industry blockchain technology, we propose a privacy protection scheme for online financial transactions based on verifiable ring signatures and blockchain by comparing and combining the unconditional anonymity provided by ring signatures with the high integrity provided by blockchain technology. Firstly, we present an algorithm for verifying ring signature based on distributed key generation, which can ensure the integrity of transaction data. Secondly, by using the block chain technique, we choose the proxy node to send the plaintext message into the block chain, and guarantee the security of the asset transaction. On this basis, the designed scheme is subjected to a security analysis to verify that it is completely anonymous, verifiable and unerasable. The protection of user privacy can be achieved while enabling online transactions. Finally, it is shown that the proposed method is more effective and practical than other similar solutions in performance assessment and simulation. It is proved that the scheme is a safe and efficient online financial transaction ring signature scheme.

Recent studies have shown that the neural network-based binary code similarity detection technology performs well in vulnerability mining, plagiarism detection, and malicious code analysis. However, existing cross-architecture methods still suffer from insufficient feature characterization and low discrimination accuracy. To address these issues, this paper proposes a cross-architecture binary function similarity detection method based on composite feature model (SDCFM). Firstly, the binary function is converted into vector representation according to the proposed composite feature model, which is composed of instruction statistical features, control flow graph structural features, and application program interface calling behavioral features. Then, the composite features are embedded by the proposed hierarchical embedding network based on a graph neural network. In which, the block-level features and the function-level features are processed separately and finally fused into the embedding. In addition, to make the trained model more accurate and stable, our method utilizes the embeddings of predecessor nodes to modify the node embedding in the iterative updating process of the graph neural network. To assess the effectiveness of composite feature model, we contrast SDCFM with the state of art method on benchmark datasets. The experimental results show that SDCFM has good performance both on the area under the curve in the binary function similarity detection task and the vulnerable candidate function ranking in vulnerability search task.

In this article, the authors focus on the use of smart CCTV, a combnation of biometric recognition technology and AI algorithms. In fact, the advancements in relevant technologies brought a significant increase in the use of biometric information - fingerprint, retina, iris or facial recognition - across diverse sectors. Both the public and private sectors, with the developments of biometric technology, widely adopt and use an individual's biometric information for different reasons. For instance, smartphone users highly count on biometric technolgies for the purpose of security. Public and private orgazanitions control an access to confidential information-controlling facilities with biometric technology. Biometric infomration is known to be unique and immutable in the course of one's life. Given the uniquness and immutability, it turned out to be as reliable means for the purpose of authentication and verification. However, the use of biometric information comes with cost, posing a privacy issue. Once it is leaked, there is little chance to recover damages resulting from unauthorized uses. The governments across the country fully understand the threat to privacy rights with the use of biometric information and AI. The EU and the United States amended their data protection laws to regulate it. South Korea aligned with them. Yet, the authors point out that Korean data aprotection law still requires more improvements to minimize a concern over privacy rights arising from the wide use of biometric information. In particular, the authors stress that it is necessary to amend Section (2) of Article 23 of PIPA to reflect the concern by changing the basis for permitting the processing of sensitive information from 'the Statutes' to 'the Acts'.

The Wireless Sensor Network (WSN), is constructed out of teeny-tiny sensor nodes that are very low-cost, have a low impact on the environment in terms of the amount of power they consume, and are able to successfully transmit data to the base station. The primary challenges that are presented by WSN are those that are posed by the distance between nodes, the amount of energy that is consumed, and the delay in time. The sensor node's source of power supply is a battery, and this particular battery is not capable of being recharged. In this scenario, the amount of energy that is consumed rises in direct proportion to the distance that separates the nodes. Here, we present a Hybrid Firefly Glow-Worm Swarm Optimization (HF-GSO) guided routing strategy for preserving WSNs' low power footprint. An efficient fitness function based on firefly optimization is used to select the Cluster Head (CH) in this procedure. It aids in minimising power consumption and the occurrence of dead sensor nodes. After a cluster head (CH) has been chosen, the Glow-Worm Swarm Optimization (GSO) algorithm is used to figure out the best path for sending data to the sink node. Power consumption, throughput, packet delivery ratio, and network lifetime are just some of the metrics measured and compared between the proposed method and methods that are conceptually similar to those already in use. Simulation results showed that the proposed method significantly reduced energy consumption compared to the state-of-the-art methods, while simultaneously increasing the number of functioning sensor nodes by 2.4%. Proposed method produces superior outcomes compared to alternative optimization-based methods.

In order to handle the problem that puncture patterns will change the position distribution of original information bits and frozen bits in polar codes, which affects performance of puncture polar codes further, a minimum row weight and polar spectrum based puncture polar codes construction algorithm (called PA-MRWP) is proposed in this paper. The algorithm calculates row weight of generator matrix and sorts the row weight in ascending order first. Next, the positions with the minimum row weight are selected as initial puncture positions. If the rows with the same row weight cannot all be punctured, polar spectrum based auxiliary puncture scheme is used. In sub-channels with the same row weight, rows corresponding to the polarized sub-channels with higher reliability are selected as puncture positions to construct puncture vector, and the reliability is calculated based on polar spectrum. It is actually a two-step selection strategy, the proposed minimum row weight puncture (MRWP) algorithm is used for primary selection and polar spectrum based auxiliary puncture is used for adjustment. Simulation results show that, compared with worst quality puncture (WQP) algorithm, the proposed PA-MRWP algorithm and Gaussian approximation-aided minimum row weight puncture (GA-MRWP) algorithm provide gains of about 0.46 dB and 0.29 dB at bit error rate (BER) of 10^-4, respectively when code length N=400, code rate R=1/2. In addition, the proposed puncture algorithms improve the BER performance significantly with respect to quasi-uniform puncture (QUP) algorithm.

The general interference is the imperative trouble for simultaneous wireless information and power transfer (SWIPT) system. Although interference has bad influences on the performance of the system, it carries energy simultaneously. In this paper, the energy-constrained relay of the SWIPT system needs to spend much time on energy collecting (EC) in the information transmission (IT) period. Therefore, we propose the scheme of interference signal energy collecting (ISEC) when the interference is strong, and the SWIPT system does not carry out IT. The relay of the system continues to collect energy and stores it until the interference has minimal impact on IT. Then the system performs IT. We divide the collected interference energy equally into several parts, and each IT block receives one part. The proposed scheme is appealing because it can reduce the time of EC in IT period to make the relay spends more time forwarding the received signal in order to improve the performance of the system throughput. Furthermore, we propose a time-switching (TS) protocol based on EC at the relay. And it allows the relay forwarding signal at an appropriate power. Under the protocol, the time of EC can be flexible according to the forwarding power that we give so that the collected energy can be used more efficiently. We give the expressions of the system throughput according to the proposed scheme and protocol. Moreover, the influence of the interference power on the system throughput is also studied.

Quality of Service (QoS) is a critical feature of Wireless Sensor Networks (WSNs) with routing algorithms. Data packets are moved between cluster heads with QoS using a number of energy-efficient routing techniques. However, sustaining high scalability while increasing the life of a WSN's networks scenario remains a challenging task. Thus, this research aims to develop an energy-balancing component that ensures equal energy consumption for all network sensors while offering flexible routing without congestion, even at peak hours. This research work proposes a Gravitational Blackhole Search Optimised splay tree routing framework. Based on the splay tree topology, the routing procedure is carried out by the suggested method using three distinct steps. Initially, the proposed GBSO decides the optimal route at initiation phases by choosing the root node with optimum energy in the splay tree. In the selection stage, the steps for energy update and trust update are completed by evaluating a novel reliance function utilising the Parent Reliance (PR) and Grand Parent Reliance (GPR). Finally, in the routing phase, using the fitness measure and the minimal distance, the GBSO algorithm determines the best route for data broadcast. The model results demonstrated the efficacy of the suggested technique with 99.52% packet delivery ratio, a minimum delay of 0.19 s, and a network lifetime of 1750 rounds with 200 nodes. Also, the comparative analysis ensured that the suggested algorithm surpasses the effectiveness of the existing algorithm in all aspects and guaranteed end-to-end delivery of packets.

As the supply of electric vehicles increases, research on wireless charging methods for convenience has been increasing. Because the electric vehicle wireless transmission device is installed on the ground and the electric vehicle battery is installed on the floor of the vehicle, the transmission and reception antennas are approximately 15-30 cm away, and thus strong magnetic fields are exposed during wireless charging. When a metallic foreign object is placed in the magnetic field area, an eddy current is induced to the metallic foreign object, and heat is generated, creating danger of fire and burns. Therefore, this study proposes a method to detect metallic foreign objects in the magnetic field area of a wireless electric vehicle charging system. An active detection-only coil array was used, and an LC resonance circuit was constructed for the frequency of the supply power signal. When a metallic foreign object is inserted into the charging zone, the characteristics of the resonance circuit are broken, and the magnitude and phase of the voltage signal at both ends of the capacitor are changed. It was confirmed that the proposed method has about 1.5 times more change than the method of comparing the voltage magnitude at one node.

Brand fandom refers to a collection of consumers with strong emotions toward a brand. Studying the dynamics of brand fandom can help brands understand which services or strategies influence their consumers to become a part of brand fandom. However, existing literature on fandom in the last three decades has mainly used qualitative methods, and there is still a lack of research on fandom using quantitative methods. Specifically, previous studies lack a framework for locating fandoms from online textual data and analyzing their dynamics. This study proposes a framework for exploring brand fandom dynamics based on online textual data. This framework consists of four phases based on the design thinking model: Preparing Data, Defining Fandom Categories, Generating Fandom Dynamics, and Analyzing Fandom Dynamics. This framework uses techniques such as social network analysis and process mining, combined with brand personality theory. We demonstrate the applicability of this framework using case studies of two Korean home appliance brands. The dataset contains 14,593 posts by consumers in 374 online communities. The results show that the proposed framework can analyze brand fandom dynamics using textual customer data. Our study contributes to the interdisciplinary research at the intersection of data-driven service design and consumer culture quantification.

This study developed digital literacy instruments and tested their effectiveness on college students' perceptions of AI technologies. In creating a new digital literacy test tool, we reviewed the concept and scale of digital literacy based on previous studies that identified the characteristics and measurement of AI literacy. We developed 23 preliminary questions for our research instrument and used a quantitative approach to survey 318 undergraduates. After conducting exploratory and confirmatory factor analysis, we found that digital literacy in the age of AI had four ability sub-factors: critical understanding, artificial intelligence social impact recognition, artificial intelligence technology utilization, and ethical behavior. Then we tested the sub-factors' predictive powers on the perception of AI's usefulness and ease of use. The regression result shows that the most common powerful predictor of the usefulness and ease of use of AI technology was the ability to use AI technology. This finding implies that for college students, the ability to use various tools based on AI technology is an essential competency in the AI era.

We propose a probabilistic fish growth model for smart aquaculture systems equipped with IoT sensors that monitor the ecological environment. As IoT sensors permeate into smart aquaculture systems, environmental data such as oxygen level and temperature are collected frequently and automatically. However, there still exists data on fish weight, tank allocation, and other factors that are collected less frequently and manually by human workers due to technological limitations. Unlike sensor data, human-collected data are hard to obtain and are prone to poor quality due to missing data and reading errors. In a situation where different types of data are mixed, it becomes challenging to develop an effective fish growth model. This study explores the unique characteristics of such a combined environmental and weight dataset. To address these characteristics, we develop a preprocessing method and a probabilistic fish growth model using mixed data sampling (MIDAS) and overlapping mixtures of Gaussian processes (OMGP). We modify the OMGP to be applicable to prediction by setting a proper prior distribution that utilizes the characteristic that the ratio of fish groups does not significantly change as they grow. We conduct a numerical study using the eel dataset collected from a real smart aquaculture system, which reveals the promising performance of our model.

Imported ornamental fish should be quarantined because they can have dangerous diseases depending on their habitat. The quarantine requires a lot of time because quarantine officers collect various information on the imported ornamental fish. Inefficient quarantine processes reduce its work efficiency and accuracy. Also, long-time quarantine causes the death of environmentally sensitive ornamental fish and huge financial losses. To improve existing quarantine systems, information on ornamental fish was collected and structured, and a server was established to develop quarantine performance support software equipped with a text search engine. However, the long names of ornamental fish in general can cause many typos and time bottlenecks when we type search words for the target fish information. Therefore, we need a technique that can correct typos. Typical typo character calibration compares input text with all characters in a calibrated candidate text dictionary. However, this approach requires computational power proportional to the number of typos, resulting in slow processing time and low calibration accuracy performance. Therefore, to improve the calibration accuracy of characters, we propose a fusion system of simple Artificial Neural Network (ANN) models and character preprocessing methods that accelerate the process by minimizing the computation of the models. We also propose a typo character generation method used for training the ANN models. Simulation results show that the proposed typo character correction system is about 6 times faster than the conventional method and has 10% higher accuracy.

With improving computing performance, various digital platforms are being developed to enable easily utilization of high-performance computing environments. EDISON 1.0 is an online simulation platform widely used in computational science and engineering education. As the research paradigm changes, the demand for developing the EDISON 1.0 platform centered on simulation into the EDISON 2.0 platform centered on data and artificial intelligence is growing. Herein, a data framework, a core module for data-centric research on EDISON 2.0 digital platform, is proposed. The proposed data framework provides the following three functions. First, it provides a data repository suitable for the data lifecycle to increase research reproducibility. Second, it provides a new data model that can integrate, manage, search, and utilize heterogeneous data to support a data-driven interdisciplinary convergence research environment. Finally, it provides an exploratory data analysis (EDA) service and data enrichment using an AI model, both developed to strengthen data reliability and maximize the efficiency and effectiveness of research endeavors. Using the EDISON 2.0 data framework, researchers can conduct interdisciplinary convergence research using heterogeneous data and easily perform data pre-processing through the web-based UI. Further, it presents the opportunity to leverage the derived data obtained through AI technology to gain insights and create new research topics.