Tech Science Press
Not a member yet
3972 research outputs found
Sort by
Splicing Image and Its Localization: A Survey
No full textWith the rapid development of information technology, digital images have become an important medium for information transmission. However, manipulating images is becoming a common task with the powerful image editing tools and software, and people can tamper the images content without leaving any visible traces of splicing in order to gain personal goal. Images are easily spliced and distributed, and the situation will be a great threat to social security. The survey covers splicing image and its localization. The present status of splicing image localization approaches is discussed along with a recommendation for future research
A Block Compressed Sensing for Images Selective Encryption in Cloud
No full textThe theory of compressed sensing (CS) has been proposed to reduce the processing time and accelerate the scanning process. In this paper, the image recovery task is considered to outsource to the cloud server for its abundant computing and storage resources. However, the cloud server is untrusted then may pose a considerable amount of concern for potential privacy leakage. How to protect data privacy and simultaneously maintain management of the image remains challenging. Motivated by the above challenge, we propose an image encryption algorithm based on chaotic system, CS and image saliency. In our scheme, we outsource the image CS samples to cloud for reduced storage and portable computing. Consider privacy, the scheme ensures the cloud to securely reconstruct image. Theoretical analysis and experiment show the scheme achieves effectiveness, efficiency and high security simultaneously
Ship Trajectory Prediction Based on BP Neural Network
No full textIn recent years, with the prosperity of world trade, the water transport industry has developed rapidly, the number of ships has surged, and ship safety accidents in busy waters and complex waterways have become more frequent. Predicting the movement of the ship and analyzing the trajectory of the ship are of great significance for improving the safety level of the ship. Aiming at the multi-dimensional characteristics of ship navigation behavior and the accuracy and real-time requirements of ship traffic service system for ship trajectory prediction, a ship navigation trajectory prediction method combining ship automatic identification system information and Back Propagation (BP) neural network are proposed. According to the basic principle of BP neural network structure, the BP neural network is trained by taking the characteristic values of ship navigation behavior at three consecutive moments as input and the characteristic values of ship navigation behavior at the fourth moment as output to predict the future ship navigation trajectory. Based on the Automatic Identification System (AIS) information of the waters near the Nanpu Bridge in Pudong New Area, Shanghai, the results show that the method is used to predict the ship's navigational behavior eigenvalues accurately and in real time. Compared with the traditional kinematics prediction trajectory method, the model can effectively predict ship navigation. The trajectory improves the accuracy of the ship's motion situation prediction, and has the advantages of high computational efficiency and strong versatility, and the error is within an acceptable range
Mathematical Analysis of Waiting Times for Reaching Therapeutic Effects
No full textIn two previous papers the authors presented mathematical models that simulate the mass of drug delivered, in vitro Ferreira, Oliveira, Silva, Carreira, Gil and Murta (2010) and in vivo Ferreira, Oliveira, Silva and Murta (2011), from a therapeutic contact lens. In the present paper the time it takes to reach an equilibrium state is studied. A closed formula based on the concept of effective time is derived and the influence of the parameters of the model is analyzed
Modelling and Backstepping Motion Control of the Aircraft Skin Inspection Robot
Full text linkAircraft skin health concerns whether the aircraft can fly safely. In this paper, an improved mechanical structure of the aircraft skin inspection robot was introduced. Considering that the aircraft skin surface is a curved environment, we assume that the curved environment is equivalent to an inclined plane with a change in inclination. Based on this assumption, the Cartesian dynamics model of the robot is established using the Lagrange method. In order to control the robot’s movement position accurately, a position backstepping control scheme for the aircraft skin inspection robot was presented. According to the dynamic model and taking into account the problems faced by the robot during its movement, a position constrained controller of the aircraft skin inspection robot is designed using the barrier Lyapunov function. Aiming at the disturbances in the robot, we adopt a fuzzy system to approximate the unknown dynamics related with system states. Finally, the simulation results of the designed position constrained controller were compared with the sliding mode controller, and prove the validity of the position constrained controller
Energy Release Rates for Interface Cracks in Multilayered Structures
Full text linkThis paper examines the evolution of the interfacial deflection energy release rates in multilayered structures under four-point bending. The J-integral and the extended finite element method (XFEM) are adopted to investigate the evolution of the interfacial deflection energy release rates of composite structures. Numerical results not only verify the accuracy of analytical solutions for the steady-state interfacial deflection energy release rate, but also provide the evolutionary history of the interfacial deflection energy release rate under different crack lengths. In addition, non-dimensional parametric analyses are performed to discuss the effects of normalized ratios of the crack length, the elastic modulus, and the thickness on the interfacial deflection energy release rate. The results demonstrate that the elastic modulus ratio and thickness ratio have a distinct influence on the interfacial deflection energy release rate for multilayered beams. Furthermore, an unstable interfacial crack tends to occur for elastic multilayer beams with higher elastic modulus on the upper sub-beam under bending moments. The unstable interfacial fracture shows a decreasing interfacial deflection energy release rate with an increasing interfacial crack length
Time-Domain Analysis of Underground Station-Layered Soil Interaction Based on High-Order Doubly Asymptotic Transmitting Boundary
Full text linkBased on the modified scale boundary finite element method and continued fraction solution, a high-order doubly asymptotic transmitting boundary (DATB) is derived and extended to the simulation of vector wave propagation in complex layered soils. The high-order DATB converges rapidly to the exact solution throughout the entire frequency range and its formulation is local in the time domain, possessing high accuracy and good efficiency. Combining with finite element method, a coupled model is constructed for time-domain analysis of underground station-layered soil interaction. The coupled model is divided into the near and far field by the truncated boundary, of which the near field is modelled by FEM while the far field is modelled by the high-order DATB. The coupled model is implemented in an open source finite element software, OpenSees, in which the DATB is employed as a super element. Numerical examples demonstrate that results of the coupled model are stable, accurate and efficient compared with those of the extended mesh model and the viscous-spring boundary model. Besides, it has also shown the fitness for long-time seismic response analysis of underground station-layered soil interaction. Therefore, it is believed that the coupled model could provide a new approach for seismic analysis of underground station-layered soil interaction and could be further developed for engineering
Defense Against Poisoning Attack via Evaluating Training Samples Using Multiple Spectral Clustering Aggregation Method
Full text linkThe defense techniques for machine learning are critical yet challenging due to the number and type of attacks for widely applied machine learning algorithms are significantly increasing. Among these attacks, the poisoning attack, which disturbs machine learning algorithms by injecting poisoning samples, is an attack with the greatest threat. In this paper, we focus on analyzing the characteristics of positioning samples and propose a novel sample evaluation method to defend against the poisoning attack catering for the characteristics of poisoning samples. To capture the intrinsic data characteristics from heterogeneous aspects, we first evaluate training data by multiple criteria, each of which is reformulated from a spectral clustering. Then, we integrate the multiple evaluation scores generated by the multiple criteria through the proposed multiple spectral clustering aggregation (MSCA) method. Finally, we use the unified score as the indicator of poisoning attack samples. Experimental results on intrusion detection data sets show that MSCA significantly outperforms the K-means outlier detection in terms of data legality evaluation and poisoning attack detection
A Neural Network-Based Trust Management System for Edge Devices in Peer-to-Peer Networks
Full text linkEdge devices in Internet of Things (IoT) applications can form peers to communicate in peer-to-peer (P2P) networks over P2P protocols. Using P2P networks ensures scalability and removes the need for centralized management. However, due to the open nature of P2P networks, they often suffer from the existence of malicious peers, especially malicious peers that unite in groups to raise each other's ratings. This compromises users' safety and makes them lose their confidence about the files or services they are receiving. To address these challenges, we propose a neural network-based algorithm, which uses the advantages of a machine learning algorithm to identify whether or not a peer is malicious. In this paper, a neural network (NN) was chosen as the machine learning algorithm due to its efficiency in classification. The experiments showed that the NNTrust algorithm is more effective and has a higher potential of reducing the number of invalid files and increasing success rates than other well-known trust management systems
New Generation Model of Word Vector Representation Based on CBOW or Skip-Gram
Full text linkWord vector representation is widely used in natural language processing tasks. Most word vectors are generated based on probability model, its bag-of-words features have two major weaknesses: they lose the ordering of the words and they also ignore semantics of the words. Recently, neural-network language models CBOW and Skip-Gram are developed as continuous-space language models for words representation in high dimensional real-valued vectors. These vector representations have recently demonstrated promising results in various NLP tasks because of their superiority in capturing syntactic and contextual regularities in language. In this paper, we propose a new strategy based on optimization in contiguous subset of documents and regression method in combination of vectors, two of new models CBOW-OR and SkipGram-OR for word vector learning are established. Experimental results show that for some words-pair, the cosine distance obtained by the CBOW-OR (or SkipGram-OR) model is generally larger and is more reasonable than CBOW (or Skip-Gram), the vector space for Skip-Gram and SkipGram-OR keep the same structure property in Euclidean distance, and the model SkipGram-OR keeps higher performance for retrieval the relative words-pair as a whole. Both CBOW-OR and SkipGram-OR model are inherent parallel models and can be expected to apply in large-scale information processing