Computing and Informatics (E-Journal - Institute of Informatics, SAS, Bratislava)
Not a member yet
1506 research outputs found
Sort by
Cellular Automata Based Image Authentication Scheme Using Extended Visual Cryptography
Get PDFMost of the Visual Cryptography based image authentication schemes hide the share and authentication data into cover images by using an additional data hiding process. This process increases the computational cost of the schemes. Pixel expansion, meaningless shares and use of codebook are other challenges in these schemes. To overcome these issues, an authentication scheme is proposed in which no embedding into the cover images is performed and meaningful authentication shares are created using the watermark and cover images. This makes the scheme completely imperceptible. The watermark can be retrieved just by superimposing these authentication shares, thus reducing the computational complexity at receiver's side. Cellular Automata is used to construct the master share that provides self-construction ability to the shares. The meaningful authentication shares help in enhancing the security of the scheme while size invariance saves transmission and storage cost. The scheme possesses the ability of tamper detection. Experimental results demonstrate the improved security and quality of the generated shares of the proposed scheme as compared to existing schemes
Perceptual Quality Assessment of Digital Images Using Deep Features
Get PDFPerceptual quality assessment is a tough task especially in the absence of reference information. No-reference image quality assessment is more challenging than full-reference or reduced reference methods, as the system has to model the different image distortions in the form of a quality score. Most of the approaches are based on handcrafted features which are based on natural scene statistics and are specific to some distortion types. These approaches provide high correlation with human opinion score for datasets containing specific distortions, but they fail to generalize well in scenarios were multiple distortions or real-time distortions are present in images. Deep learning algorithms, on the other hand, demonstrated their abilities to learn expert features with better discriminatory power for various classification and regression tasks. It is a big challenge to use those deep learning methods for image quality assessment as the image datasets with human opinion score are very small and cannot be used effectively to train a deep learning algorithm. We experimented with activations of different deep layers of thirteen pre-trained models and checked for their suitability for the task of no-reference quality assessment. Fine-tuning of these models on quality assessment datasets provided even better performance. A Gaussian process regression model is trained on these activations to perform the quality assessment and it provided state-of-the-art performance. Cross-dataset validation demonstrated its performance further and also provided further prospects of research in this direction
An Improved PDR Localization Algorithm Based on Particle Filter
Get PDFPedestrian Dead Reckoning (PDR) helps to realize step frequency detection, step estimation and direction estimation through data collected by inertial sensors such as accelerometer, gyroscope, magnetometer, etc. The initial positioning information is used to calculate the position of pedestrians at any time, which can be applied to indoor positioning technology researching. In order to improve the position accuracy of pedestrian track estimation, this paper improves the step frequency detection, step size estimation and direction detection in PDR, and proposes a particle swarm optimization particle filter (PSO-IPF) PDR location algorithm. Using the built-in accelerometer information of the smartphone to carry out the step frequency detection, the step frequency parameter construction model is introduced to carry out the step estimation, the direction estimation is performed by the Kalman filter fusion gyroscope and the magnetometer information, and the positioning data is merged by using the particle filter. The fitness function in the particle swarm optimization process is changed in the localization algorithm to improve particle diversity and position estimation. The experimental results show that the error rate of the improved step frequency detection method is reduced by about 2.1 % compared with the traditional method. The angle accuracy of the direction estimation is about 4.12◦ higher than the traditional method. The overall positioning accuracy is improved
Traffic Light Recognition for Real Scenes Based on Image Processing and Deep Learning
Get PDFTraffic light recognition in urban environments is crucial for vehicle control. Many studies have been devoted to recognizing traffic lights. However, existing recognition methods still face many challenges in terms of accuracy, runtime and size. This paper presents a novel robust traffic light recognition approach that takes into account these three aspects based on image processing and deep learning. The proposed approach adopts a two-stage architecture, first performing detection and then classification. In the detection, the perspective relationship and the fractal dimension are both considered to dramatically reduce the number of invalid candidate boxes, i.e. region proposals. In the classification, the candidate boxes are classified by SqueezeNet. Finally, the recognized traffic light boxes are reshaped by postprocessing. Compared with several reference models, this approach is significantly competitive in terms of accuracy and runtime. We show that our approach is lightweight, easy to implement, and applicable to smart terminals, mobile devices or embedded devices in practice
New Software Tool for Modelling and Control of Discrete-Event and Hybrid Systems Using Petri Nets
Get PDFThe main aim of the proposed paper is to design a new software tool for modelling and control of discrete-event and hybrid systems using Arduino and similar microcontrollers. To accomplish these tasks a new tool called PN2ARDUINO based on Petri nets is proposed which is able to communicate with the microcontroller. Communication with the microcontroller is based on the modified Firmata protocol hence the control algorithm can be implemented on all microcontrollers that support this type of protocol. The developed software tool has been successfully verified in control of laboratory systems. It can also be used for education and research purposes as it offers a graphical environment for designing control algorithms for hybrid and mainly discrete-event systems. The proposed tool can improve education and practice in the field of cyber-physical systems (Industry 4.0)
Coalgebraic Operational Semantics for an Imperative Language
Get PDFOperational semantics is a known and popular semantic method for describing the execution of programs in detail. The traditional definition of this method defines each step of a program as a transition relation. We present a new approach on how to define operational semantics as a coalgebra over a category of configurations. Our approach enables us to deal with a program that is written in a small but real imperative language containing also the common program constructs as input and output statements, and declarations. A coalgebra enables to define operational semantics in a uniform way and it describes the behavior of the programs. The state space of our coalgebra consists of the configurations modeling the actual states; the morphisms in a base category of the coalgebra are the functions defining particular steps during the program's executions. Polynomial endofunctor determines this type of systems. Another advantage of our approach is its easy implementation and graphical representation, which we illustrate on a simple program
Analysis and Application of Min-Cost Transition Systems to Business Process Management
Get PDFTo improve the efficiency of conformance checking in process mining, new alignment approaches are presented between event logs and process models based on the min-cost transition systems of Petri nets. An algorithm is presented to obtain the transition system with the minimum cost based on the product of the event net and process net. The min-cost transition system is a directed acyclic graph, where the paths from the initial node to the final node include all optimal alignments between the trace and the process model based on the given cost function. Two algorithms are proposed to calculate an optimal alignment and all optimal alignments, respectively. All algorithms are implemented in ProM platform. After a series of the simulation experiments, the feasibility and effectiveness of the proposed approaches are illustrated
Gathering Information on the Web by Consistent Entity Augmentation
Get PDFUsers usually want to gather information about what they are interested in, which could be achieved by entity augmentation using a vast amount of web tables. Existing techniques assume that web tables are entity-attribute binary tables. As for tables having multiple columns to be augmented, they will be split into several entity-attribute binary relations, which would cause semantic fragmentation. Furthermore, the result table consolidated by binary relations will suffer from entity inconsistency and low precision. The objective of our research is to return a consistent result table for entity augmentation when given a set of entities and attribute names. In this paper we propose a web information gathering framework based on consistent entity augmentation. To ensure high consistency and precision of the result table we propose that answer tables for building result table should have consistent matching relationships with each other. Instead of splitting tables into pieces we regard web tables as nodes and consistent matching relationships as edges to make a consistent clique and expand it until its coverage for augmentation query reaches certain threshold gamma. It is proved in this paper that a consistent result table could be built by considering tables in consistent clique to be answer tables. We tested our method on four real-life datasets, compared it with different answer table selection methods and state-of-the-art entity augmentation technique based on table fragmentation as well. The results of a comprehensive set of experiments indicate that our entity augmentation framework is more effective than the existing method in getting consistent entity augmentation results with high accuracy and reliability
New Approach to Edge Detection on Different Level of Wavelet Decomposition
Get PDFThis paper proposes a new approach to edge detection on the images over which the wavelet decomposition was done to the third level and consisting of different levels of detail (small, medium and high level of detail). Images from the BSD (Berkeley Segmentation Dataset) database with the corresponding ground truth were used. Daubechies wavelet was used from second to tenth order. Gradient and Laplacian operators were used for edge detection. The proposed approach is applied in systems where information is processed in real time, where fast image processing is required and in systems where high compression ratio is used. That is, it can find practical application in many systems, especially in television systems where the level of details in the image changes. The new approach consists in the fact that when wavelet transform is applied, an edge detection is performed over the level 1 image to create a filter. The filter will record only those pixels that can be potential edges. The image is passed through a median filter that filters only the recorded pixels and 8 neighbors of pixel. After that, the edge detection with one of the operators is applied onto the filtered image. F measure, FoM (Figure of Merit) and PR (Performance Ratio) were used as an objective measure. Based on the obtained results, the application of the proposed approach achieves significant improvements and these improvements are very good depending on the number of details in the image and the compression ratio. These results and improvements can be used to improve the quality of edge detection in many systems where compressed images are processed, that is, where work with images with a high compression ratio is required
Virtual Machine Deployment Strategy Based on Improved PSO in Cloud Computing
Get PDFEnergy consumption is an important cost driven by growth of computing power, thereby energy conservation has become one of the major problems faced by cloud system. How to maximize the utilization of physical machines, reduce the number of virtual machine migrations, and maintain load balance under the constraints of physical machine resource thresholds that is the effective way to implement energy saving in data center. In the paper, we propose a multi-objective physical model for virtual machine deployment. Then the improved multi-objective particle swarm optimization (TPSO) is applied to virtual machine deployment. Compared to other algorithms, the algorithm has better ergodicity into the initial stage, improves the optimization precision and optimization efficiency of the particle swarm. The experimental results based on CloudSim simulation platform show that the algorithm is effective at improving physical machine resource utilization, reducing resource waste, and improving system load balance