Computing and Informatics (E-Journal - Institute of Informatics, SAS, Bratislava)
Not a member yet
1506 research outputs found
Sort by
Low-Light Image Enhancement via Weighted Fractional-Order Model
Full text linkLow-light image enhancement (LLIE) enables to serve high-level vision tasks and improve their efficiency. Retinex-based methods have well been recognized as a representative technique for LLIE, but they still suffer from inflexible regularization terms in decomposing illumination and reflectance. In this paper, we propose a new weighted fractional-order variational model based on the Retinex model. First, the constructed weighted fractional-order variational model estimates piecewise smoothed and weakly pixel-shifted illumination by aware structures and textures. Then, to solve this problem accurately, we chose a semi-decoupled approach and an alternating minimization method. Finally, the designed multi-illumination fusion method accurately enhances the structure-rich dark regions of the image through well-exposedness and local entropy weights, while realizing adaptive enhancement based on a naturalness-preserving parameter estimation algorithm. The results of subjective and objective experiments on several challenging low-light datasets demonstrate that our proposed method shows better competitiveness in enhancing low-light images compared with the state-of-the-art methods
Research on Discrimination Method of Carbon Deposit Degree of Automobile Engine Based on Deep Learning
Full text linkThe detection of carbon deposit degree is of great significance to the maintenance of automobile engine. Due to issues with poor feature aggregation, inter-class similarity, and intra-class variance in carbon deposit data with a small number of samples, model-based discriminative approaches cannot be widely implemented in the market. In order to overcome this technical barrier, the article examines the impact of DCNNs (Deep Convolutional Neural Networks) level on the recognition effect of the degree of carbon deposit, introduces a dropout structure and data enhancement strategy to lower the risk of overfitting brought on by the small dataset, and suggests a recognition method based on the kernel of dual-dimensional multiscale-multifrequency information features to enhance the differentiation characteristic. After experimental testing, the accuracy of this method is 86.9 %, the F1-score is 87.2 %, and the inference speed is 190 FPS, which can meet the practical requirements and provide basic support for the large-scale promotion of the model discrimination
Multiple Features Extraction and Fusion for Ultrasound Dynamic Images Classification
Full text linkUltrasound examination is of great significance in the clinical diagnosis of diseases. Processing and analyzing ultrasound images through artificial intelligence technology and providing assistance in decision-making has been a hot topic of research for several years. However, since most medical images exist in the form of pictures, the current processing methods for ultrasound images basically continue to adopt the technical achievements related to static medical image processing not considering the characteristics reflected by the dynamically changing ultrasound images thus resulting in a missed diagnosis of diseases. To this end, this paper proposes an innovative multi-feature extraction and fusion method for ultrasound dynamic image classification which first extracts various types of underlying features such as texture, edge, and shape of salient targets in medical images that apply to dynamic images. Then, the feature frequency-inverse image frequency (FF-IIF) multi-feature fusion algorithm is used to generate an adaptive combined feature classification. In the experiments, the effects of the proposed algorithm are verified for three ultrasound examination items respectively. The experimental results show that the features extracted by the multi-feature fusion algorithm using FF-IIF still maintain a certain degree of fault tolerance and stability under the dynamic changes of ultrasound probe position and orientation. The computation time of the algorithm is moderate and perfectly adapted to the real-time examination of ultrasound medicine
Application and Effectiveness of IoT Edge and Fog Computing Technologies in Smart Energy Development with the Use of Encryption Algorithms and Security Systems
Full text linkThe invention of new technologies and the upgrading of existing ones to carry out assigned duties for human requirements efficiently and quickly are among the new prospects made possible by the evolution of artificial intelligence and information technology. This study aims to evaluate the performance of the Internet of Things solutions based on fog and edge computing. This article examines fog computing architectures and discusses the primary potential security and trust issues. The topical issue is to analyze and improve the security of data transmission from electronic devices and media in fog clouds using data encryption and encryption algorithms. The study confirms that depending on the amount of information to be processed, each block encryption algorithm can have a different key length. The study investigates the Advanced Encryption Standard (AES) algorithm as a more typical and efficient block encryption algorithm with key lengths of 128 and 256 bits. The paper contains a comparative characteristic of data processing speed with the size of 512 MB on the example of the AES and PRESENT block encryption algorithms, demonstrating an effective software implementation. The main scientific novelty of the study is that the efficiency of such coding algorithms as AES and PRESENT increases with a decrease in their size (from 256 to 128-bit and from 128 to 80-bit, respectively). It is necessary to expand our results for a more complex residential system
Load Forecasting of Sparrow Search Algorithm Optimization Double BIGRU
Full text linkIn this paper, a PCA-SSA-DBIGRU-Attention multi-factor short-term power load forecasting model is proposed. Taking a complete account of the influence of meteorological factors, principal components analysis (PCA) is used to analyze the meteorological factors of daily minimum, maximum, daily average temperature, relative humidity, daily precipitation and power load data at the same time. The realization of original load data is dimensioned down. The complexity of power load forecasting models is reduced. Then, the Attention Double Bidirectional Gating Recurrent Unit (DBIGRU) model is constructed to calculate the different weights of the hidden layer states of the two-layer BIGRU. The hidden layer states are assigned different weights. The Sparrow Search Algorithm (SSA) is incorporated into the DBIGRU-Attention. The SSA-DBIGRU-Attention network model is constructed to optimize the learning rate, the number of iterations and the four hyperparameters of the first and second hidden layer neurons. The extracted principal components are input into SSA-DBIGRU-Attention to realize multi-factor short-term power load forecasting. Experimental results show that the prediction accuracy of the proposed model is improved, and the prediction time is reduced. Compared to the VMD-BILSTM, PCA-DBILSTM, CNN-GRU-Attention and CNN-BIGRU-Attention model, the four aspects of MAPE, MAE, RMSE and time are reduced by 29.55 %, 36.42 %, 32.34 % and 12.22 %, respectively, the R2 is improved by 3.09 %
Micro-Directional Propagation Method Based on User Clustering
Full text linkWith the development of recommendation technology, it is of great significance to analyze users' digital footprints on social networking sites, extract user behavior rules, and make a relatively accurate assessment of each user's needs, to provide personalized services for users. It has been found that the users' behavior on social networking sites has a great correlation with the user's personalities. The OCEAN model's five characteristics can cover all aspects of user personality. There are some shortcomings in the current micro-directional propagation method. This paper has improved the traditional collaborative filtering method and proposed a collaborative filtering method based on user clustering. The model first clusters the users according to their OCEAN model, and then it filters the users collaboratively in the cluster to which the user belongs with the collaborative filtering method based on an optimized singular value decomposition (SVD) recommendation algorithm, called the BiasSVD recommendation algorithm -- to reduce the dimensionality of the data. Then it generates recommendations. Experiments show that clustering users' OCEAN models can improve the accuracy of recommendations. Compared with the entire user space, it greatly reduces the recommendation time, and effectively solves the cold start problem in micro directional propagation
HS-CGK: A Hybrid Sampling Method for Imbalance Data Based on Conditional Tabular Generative Adversarial Network and K-Nearest Neighbor Algorithm
Full text linkClass imbalance problem in datasets can lead to biased classification decisions in favor of majority class samples. Additionally, class overlap can cause fuzzy classification boundaries, affecting the performance of classification algorithms. To address these issues, we propose a hybrid sampling method based on conditional tabular generative adversarial network (CTGAN) and K-nearest neighbor (KNN) algorithm. Firstly, we introduce an oversampling algorithm, named DB-CTGAN, based on CTGAN. This algorithm filters noisy and boundary samples using the density-based spatial clustering of applications with noise (DBSCAN) clustering algorithm and generates synthetic samples that conform to the real data distribution using CTGAN. Finally, we combine the expanded fraudulent samples generated by DB-CTGAN with the normal samples and use the KNN overlap undersampling algorithm to remove the samples in the overlap region, solving the class overlap problem. Experimental results show that compared with eight sampling methods using four standard classification models (Random Forest, Decision Tree, Support Vector Classification, and XGBoost), the proposed method significantly improves the F1, AUC, and G-mean metrics on five real datasets
Exploring the Impact of Security-Based Non-Functional Requirements on Early Software Size Estimation
Full text linkSoftware size -- often measured in the source lines of code (SLOC) -- fundamentally determines the software development effort. Realistic estimates of software size are, therefore, crucial for project planning. However, even though software size is influenced by both functional requirements (FRs) and non-functional requirements (NFRs), NFRs have been largely neglected in previous SLOC estimation studies. This study conducts an initial investigation of the impact of NFRs on early SLOC estimation by focusing on security-based NFRs related to data entry validations. First, the IFPUG software non-functional assessment process (SNAP) is used to calculate SNAP points for data entry validations (SPDEV). Then, SPDEV is used along with specially adjusted analysis class diagram (ACD) metrics to build and validate an early SLOC estimation model using an industrial dataset. Finally, the proposed model is compared with two existing size estimation models. Results indicate that our proposed model outperforms both models in terms of estimation accuracy
Travel Interest Point Recommendation Algorithm Based on Collaborative Filtering and Graph Convolutional Neural Networks
Full text linkTourist attraction recommendation algorithms have been developed to meet demand related to tourism, spiritual and cultural pursuits. While many studies have been conducted on such algorithms, problems remain regarding tourist interest point recommendation such as ignoring social information, underutilizing context information, and not capturing node relationships which have limited the recommendation performance and representation capability. This paper proposes an algorithm based on graph convolutional neural networks and collaborative filtering (GCNs-CF) for travel interest point recommendation, using an image denoising encoder (IDE) instead of domain aggregation, to better capture the relationships and features between users and adjacent nodes of travel interest point nodes. An adaptive adjustment of the negative sample gradient size is used to solve the problem of slow convergence of graph convolutional neural network. The experimental results show that the proposed method has a higher recommendation effect than other algorithms
FESNet: Spotting Facial Expressions Using Local Spatial Discrepancy and Multi-Scale Temporal Aggregation
Full text linkFacial expressions (FEs) spotting aims to split long videos into intervals of neutral expression, macro-expression, or micro-expression. Recent works mainly focus on feature descriptor or optical flow methods, suffering from difficulty capturing subtle facial motion and efficient temporal aggregation. This paper proposes a novel end-to-end network, named FESNet (Facial Expression Spotting Network), to solve the above challenges. The main idea is to model the subtle facial motion as local spatial discrepancy and incorporate temporal correlation by multi-scale temporal convolution. The FESNet comprises a local spatial discrepancy module (LSDM) and a multi-scale temporal aggregation module (MTAM). The LSDM first extracts the static spatial features from each frame by residual convolution and learns the inner spatial correlation by multi-head attention. Moreover, the subtle facial motion of facial expression is modeled as the discrepancy between the first frame and the current frame of the input interval, making frame-wise spatial proposals. Using the local spatial discrepancy features and proposals as input, the MTAM incorporates the temporal correlation by multi-scale temporal convolution and performs cascade refinement to make the final prediction. Furthermore, this paper proposes a smooth loss to ensure the temporal consistency of the cascade refined proposals from MTAM. Comprehensive experiments show that FESNet achieves competitive performance compared to state-of-the-art methods