With the unprecedented growth of textual information on the Internet, an efficient automatic summarization system has become an urgent need. Recently, the neural network models based on the encoder-decoder with an attention mechanism have demonstrated powerful capabilities in the sentence summarization task. However, for paragraphs or longer document summarization, these models fail to mine the core information in the input text, which leads to information loss and repetitions. In this paper, we propose an abstractive document summarization method by applying guidance signals of key sentences to the encoder based on the hierarchical encoder-decoder architecture, denoted as KI-HABS. Specifically, we first train an extractor to extract key sentences in the input document by the hierarchical bidirectional GRU. Then, we encode the key sentences to the key information representation in the sentence level. Finally, we adopt key information representation guided selective encoding strategies to filter source information, which establishes a connection between the key sentences and the document. We use the CNN/Daily Mail and Gigaword datasets to evaluate our model. The experimental results demonstrate that our method generates more informative and concise summaries, achieving better performance than the competitive models.

For the research field of night infrared surveillance video, the target imaging in the video is easily affected by the light due to the characteristics of the active infrared camera and the classical ViBe algorithm has some problems for moving target extraction because of background misjudgment, noise interference, ghost shadow and so on. Therefore, an improved ViBe algorithm (I-ViBe) for moving target extraction in night infrared surveillance video is proposed in this paper. Firstly, the video frames are sampled and judged by the degree of light influence, and the video frame is divided into three situations: no light change, small light change, and severe light change. Secondly, the ViBe algorithm is extracted the moving target when there is no light change. The segmentation factor of the ViBe algorithm is adaptively changed to reduce the impact of the light on the ViBe algorithm when the light change is small. The moving target is extracted using the region growing algorithm improved by the image entropy in the differential image of the current frame and the background model when the illumination changes drastically. Based on the results of the simulation, the I-ViBe algorithm proposed has better robustness to the influence of illumination. When extracting moving targets at night the I-ViBe algorithm can make target extraction more accurate and provide more effective data for further night behavior recognition and target tracking.

The importance and necessity of artificial intelligence, particularly machine learning, has recently been emphasized. In fact, artificial intelligence, such as intelligent surveillance cameras and other security systems, is used to solve various problems or provide convenience, providing solutions to problems that humans traditionally had to manually deal with one at a time. Among them, information security is one of the domains where the use of artificial intelligence is especially needed because the frequency of occurrence and processing capacity of dangerous codes exceeds the capabilities of humans. Therefore, this study intends to examine the definition of artificial intelligence and machine learning, its execution method, process, learning algorithm, and cases of utilization in various domains, particularly the cases and contents of artificial intelligence technology used in the field of information security. Based on this, this study proposes a method to apply machine learning technology to the method of classifying and detecting malware that has rapidly increased in recent years. The proposed methodology converts software programs containing malicious codes into images and creates training data suitable for machine learning by preparing data and augmenting the dataset. The model trained using the images created in this manner is expected to be effective in classifying and detecting malware.

The SinGAN is one of generative adversarial networks that can be trained on a single nature image. It has poor ability to learn more global features from nature image, and losses much local detail information when it generates arbitrary size image sample. To solve the problem, a non-linear function is firstly proposed to control downsampling ratio that is ratio between the size of current image and the size of next downsampled image, to increase the ratio with increase of the number of downsampling. This makes the low-resolution images obtained by downsampling have higher proportion in all downsampled images. The low-resolution images usually contain much global information. Therefore, it can help the model to learn more global feature information from downsampled images. Secondly, the attention mechanism is introduced to the generative network to increase the weight of effective image information. This can make the network learn more local details. Besides, in order to make the output image more natural, the TVLoss function is introduced to the loss function of SinGAN, to reduce the difference between adjacent pixels and smear phenomenon for the output image. A large number of experimental results show that our proposed model has better performance than other methods in generating random samples with fixed size and arbitrary size, image harmonization and editing.

Deep Learning as a Service (DLaaS), utilizing the cloud-based deep neural network models to provide customer prediction services, has been widely deployed on mobile cloud computing (MCC). Such services raise privacy concerns since customers need to send private data to untrusted service providers. In this paper, we devote ourselves to building an efficient protocol to classify users' images using the convolutional neural network (CNN) model trained and held by the server, while keeping both parties' data secure. Most previous solutions commonly employ homomorphic encryption schemes based on Ring Learning with Errors (RLWE) hardness or two-party secure computation protocols to achieve it. However, they have limitations on large communication overheads and costs in MCC. To address this issue, we present LeHE4SCNN, a scalable privacy-preserving and communication-efficient framework for CNN-based DLaaS. Firstly, we design a novel low-expansion rate homomorphic encryption scheme with packing and unpacking methods (LeHE). It supports fast homomorphic operations such as vector-matrix multiplication and addition. Then we propose a secure prediction framework for CNN. It employs the LeHE scheme to compute linear layers while exploiting the data shuffling technique to perform non-linear operations. Finally, we implement and evaluate LeHE4SCNN with various CNN models on a real-world dataset. Experimental results demonstrate the effectiveness and superiority of the LeHE4SCNN framework in terms of response time, usage cost, and communication overhead compared to the state-of-the-art methods in the mobile cloud computing environment.

The dynamic network state and the mobility of the terminals make the service function chain (SFC) orchestration mechanisms based on static and deterministic assumptions hard to be applied in SDN/NFV mobile edge computing networks. Designing dynamic and online SFC orchestration mechanism can greatly improve the execution efficiency of compute-intensive and resource-hungry applications in mobile edge computing networks. In order to increase the overall profit of service provider and reduce the resource cost, the system running time is divided into a sequence of time slots and a dynamic orchestration scheme based on an improved column generation algorithm is proposed in each slot. Firstly, the SFC dynamic orchestration problem is formulated as an integer linear programming (ILP) model based on layered graph. Then, in order to reduce the computation costs, a column generation model is used to simplify the ILP model. Finally, a two-stage heuristic algorithm based on greedy strategy is proposed. Four metrics are defined and the performance of the proposed algorithm is evaluated based on simulation. The results show that our proposal significantly provides more than 30% reduction of run time and about 12% improvement in service deployment success ratio compared to the Viterbi algorithm based mechanism.

The Internet of Things (IoT) eHealth systems composed by Wireless Body Area Network (WBAN) has emerged recently. Sensor nodes are placed around or in the human body to collect physiological data. WBAN has many different applications, for instance health monitoring. Since the limitation of the size of the battery, besides speed, reliability, and accuracy; design of WBAN protocols should consider the energy efficiency and time delay. To solve these problems, this paper adopt the end-edge-cloud orchestrated network architecture and propose a transmission based on reinforcement algorithm. The priority of sensing data is classified according to certain application. System utility function is modeled according to the channel factors, the energy utility, and successful transmission conditions. The optimization problem is mapped to Q-learning model. Following this online power control protocol, the energy level of both the senor to coordinator, and coordinator to edge server can be modified according to the current channel condition. The network performance is evaluated by simulation. The results show that the proposed power control protocol has higher system energy efficiency, delivery ratio, and throughput.

In the 4th Industrial Revolution, the healthcare industry is undergoing a paradigm shift from post-care and management systems based on diagnosis and treatment to disease prevention and management based on personal precision medicine. To optimize medical services for individual patients, an open ecosystem for the healthcare industry that allows the exchange and utilization of personal health records (PHRs) is required. However, under the current system of hospital-centered data management, it is difficult to implement the linking and sharing of PHRs in practice. To address this problem, in this study, we present the design and implementation of a patient-centered PHR platform using blockchain technology. This platform achieved transparency and reliability in information management by eliminating the risk of leakage and tampering/altering personal information, which could occur when using a PHR. In addition, the patient-consent system was applied to a PHR; thus, the patient acted as the user with ownership. The proposed blockchain-based PHR platform enables the integration of personal medical information with scattered distribution across multiple hospitals, and allows patients to freely use their health records in their daily lives and emergencies. The proposed platform is expected to serve as a stepping stone for patient-centered healthcare data management and utilization.

Image dehazing is an ill-posed problem which is far from being solved. Traditional image dehazing methods often yield mediocre effects and possess substandard processing speed, while modern deep learning methods perform best only in certain datasets. The haze removal effect when processed by said methods is unsatisfactory, meaning the generalization performance fails to meet the requirements. Concurrently, due to the limited processing speed, most dehazing algorithms cannot be employed in the industry. To alleviate said problems, a lightweight fast dehazing network based on a multiple scale-patch framework (MSP) is proposed in the present paper. Firstly, the multi-scale structure is employed as the backbone network and the multi-patch structure as the supplementary network. Dehazing through a single network causes problems, such as loss of object details and color in some image areas, the multi-patch structure was employed for MSP as an information supplement. In the algorithm image processing module, the image is segmented up and down for processed separately. Secondly, MSP generates a clear dehazing effect and significant robustness when targeting real-world homogeneous and nonhomogeneous hazy maps and different datasets. Compared with existing dehazing methods, MSP demonstrated a fast inference speed and the feasibility of real-time processing. The overall size and model parameters of the entire dehazing model are 20.75M and 6.8M, and the processing time for the single image is 0.026s. Experiments on NTIRE 2018 and NTIRE 2020 demonstrate that MSP can achieve superior performance among the state-of-the-art methods, such as PSNR, SSIM, LPIPS, and individual subjective evaluation.

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

The system for re-identifying persons is used to find and verify the persons crossing through different spots using various cameras. Much research has been done to re-identify the person by utilising features with deep-learned or hand-crafted information. Deep learning techniques segregate and analyse the features of their layers in various forms, and the output is complex feature vectors. This paper proposes a distinctive framework called Integrated Level Feature Pattern (ILFP) framework, which integrates local and global features. A new deep learning architecture named modified XceptionNet (m-XceptionNet) is also proposed in this work, which extracts the global features effectively with lesser complexity. The proposed framework gives better performance in Rank1 metric for Market1501 (96.15%), CUHK03 (82.29%) and the newly created NEC01 (96.66%) datasets than the existing works. The mean Average Precision (mAP) calculated using the proposed framework gives 92%, 85% and 98%, respectively, for the same datasets.

Video service providers tend to face user network problems in the process of transmitting video streams. They strive to provide user with superior video quality in a limited bitrate environment. It is necessary to accurately determine the target bitrate range of the video under different quality requirements. Recently, several schemes have been proposed to meet this requirement. However, they do not take the impact of visual influence into account. In this paper, we propose a new multi-category model to accurately predict the target bitrate range with target visual quality by machine learning. Firstly, a dataset is constructed to generate multi-category models by machine learning. The quality score ladders and the corresponding bitrate-interval categories are defined in the dataset. Secondly, several types of spatial-temporal features related to VMAF evaluation metrics and visual factors are extracted and processed statistically for classification. Finally, bitrate prediction models trained on the dataset by RandomForest classifier can be used to accurately predict the target bitrate of the input videos with target video quality. The classification prediction accuracy of the model reaches 0.705 and the encoded video which is compressed by the bitrate predicted by the model can achieve the target perceptual quality.

In camera sensor networks (CSNs), in order to better identify the point, full-view problem requires capture any facing direction of target (point or intruder), and its coverage prediction and sensor density issues are more complicated. At present, a lot of research supposes that a large number of homogeneous camera sensors are randomly distributed in a bounded square monitoring region to obtain full-view rate which is close to 1. In this paper, we deduce the sensor density prediction model in heterogeneous deployed CSNs with arbitrary full-view rate. Aiming to reduce the influence of boundary effect, we introduce the concepts of expanded monitoring region and maximum detection area. Besides, in order to verify the performance of the proposed sensor density model, we carried out different scenarios in simulation experiments to verify the theoretical results. The simulation results indicate that the proposed model can effectively predict the sensor density with arbitrary full-view rate.

Today distributed data storage service are being widely used. However lack of proper means of security makes the user data vulnerable. In this work, we propose a Randomized Block Size (RBS) model for secure data storage in distributed environments. The model work with multifold block sizes encrypted with the Chinese Remainder Theorem-based RSA (C-RSA) technique for end-to-end security of multimedia data. The proposed RBS model has a key generation phase (KGP) for constructing asymmetric keys, and a rand generation phase (RGP) for applying optimal asymmetric encryption padding (OAEP) to the original message. The experimental results obtained with text and image files show that the post encryption file size is not much affected, and data is efficiently encrypted while storing at the distributed storage server (DSS). The parameters such as ciphertext size, encryption time, and throughput have been considered for performance evaluation, whereas statistical analysis like similarity measurement, correlation coefficient, histogram, and entropy analysis uses to check image pixels deviation. The number of pixels change rate (NPCR) and unified averaged changed intensity (UACI) were used to check the strength of the proposed encryption technique. The proposed model is robust with high resilience against eavesdropping, insider attack, and chosen-plaintext attack.

Because the traditional network information security vulnerability risk assessment method does not set the weight, it is easy for security personnel to fail to evaluate the value of information security vulnerability risk according to the calculation value of network centrality, resulting in poor evaluation effect. Therefore, based on the network security data element feature system, this study designed a quantitative assessment method of network information security vulnerability detection risk under single transmission state. In the case of single transmission state, the multi-dimensional analysis of network information security vulnerability is carried out by using the analysis model. On this basis, the weight is set, and the intrinsic attribute value of information security vulnerability is quantified by using the qualitative method. In order to comprehensively evaluate information security vulnerability, the efficacy coefficient method is used to transform information security vulnerability associated risk, and the information security vulnerability risk value is obtained, so as to realize the quantitative evaluation of network information security vulnerability detection under single transmission state. The calculated values of network centrality of the traditional method and the proposed method are tested respectively, and the evaluation of the two methods is evaluated according to the calculated results. The experimental results show that the proposed method can be used to calculate the network centrality value in the complex information security vulnerability space network, and the output evaluation result has a high signal-to-noise ratio, and the evaluation effect is obviously better than the traditional method.

The evolution of IoT technology is having a significant impact on people's lives. Almost all areas of people's lives are benefiting from increased productivity and simplification made possible by this trending technology. On the downside, however, the application of IoT technology is posing some security challenges, among them, unauthorized access to IoT devices. This paper presents an Attribute-based Access Control Fog architecture that aims to achieve effective distribution, increase availability and decrease latency. In the proposed architecture, the main functional points of the Attribute-based Access Control are distributed to provide policy decision and policy information mechanisms in fog nodes, locating these functions near end nodes. To evaluate the proposed architecture, an access control engine based on the Attribute-based Access Control was built using the Balana library and simulated using EdgeCloudSim to compare it to the traditional cloud-based architecture. The experiments show that the fog-based architecture provides robust results in terms of reducing latency in making access decisions.

This study proposes an analytical approximation algorithm based on extreme value theory (EVT) for the inverse of the power of the incomplete Gamma function. First, the Gumbel function is used to approximate the power of the incomplete Gamma function, and the corresponding inverse problem is transformed into the inversion of an exponential function. Then, using the tail equivalence theorem, the normalized coefficient of the general Weibull distribution function is employed to replace the normalized coefficient of the random variable following a Gamma distribution, and the approximate closed form solution is obtained. The effects of equation parameters on the algorithm performance are evaluated through simulation analysis under various conditions, and the performance of this algorithm is compared to those of the Newton iterative algorithm and other existing approximate analytical algorithms. The proposed algorithm exhibits good approximation performance under appropriate parameter settings. Finally, the performance of this method is evaluated by calculating the thresholds of space-time block coding and space-frequency block coding pattern recognition in multiple-input and multiple-output orthogonal frequency division multiplexing. The analytical approximation method can be applied to other related situations involving the maximum statistics of independent and identically distributed random variables following Gamma distributions.

In IEEE 802.11p-based wireless access in vehicular environments (WAVE) communication systems, channel estimation (CE) is one of the important issues to provide stable communication service. It is hard to apply conventional CE schemes based on data pilot to real systems, because error propagation occurs in high mobility and modulation order environments, resulting in degrading the CE performance. In this paper, we propose one data pilot using multiple receive antennas (ODPM) CE scheme based on the weighted sum using update matrix (WSUM) with time-domain averaging (TDA) to overcome this problem. Within the process of WSUM-TDA in the proposed scheme, the maximum ratio combining (MRC) technique is applied so as to create more accurate one data pilot. Moreover, a new reliability test criterion is proposed as the fashion of utilizing MRC, which makes it possible to apply selective TDA that guarantees performance improvement. In simulation results, the packet error rate (PER) performance of the proposed ODPM is compared with that of conventional CE methods and its superiority is demonstrated.

In cognitive radio networks where unlicensed secondary users opportunistically access to licensed spectrum unused by licensed primary users, spectrum sensing is one of the key issues in order to effectively use the frequency resource. For enhancing the sensing performance in energy detection-based spectrum sensing, spatial diversity based on multiple antennas is utilized. However, the sensing performance can be degraded when antennas are spatially correlated, resulting in inducing the harmful interference to primary users. To overcome this problem, in this paper, an advanced energy detector is proposed. In the proposed sensing method, a weight matrix based on the eigenvalues of the spatial channels without any prior information on the primary signals is defined and utilized. In numerical simulations, it is shown that the proposed detector outperforms the conventional detector with regard to false-alarm and detection probabilities when antenna are spatially correlated.

IEEE 802.11ax is a protocol being developed for high-density Wireless Local Area Networks (WLAN). Several algorithms have been proposed to improve the level of spatial reuse applied in IEEE 802.11ax. However, these algorithms are tentative and do not specify how to select the transmit power and carrier sense threshold in practice; It is unclear when and why the tuned parameters lead to better network performance. In this paper, we restricted the scale of transmit power tuning to prevent the case of backfire in which spatial reuse will result in transmission failure. If the restrictions cannot be satisfied, spatial reuse will be abandoned. This is why we named the proposed scheme as Arbitration based Spatial Reuse (ASR). We quantified the network performance after spatial reuse, and formulate a corresponding maximum problem whose solution is the optimal carrier sense threshold and transmit power. We verified our theoretical analysis by simulation and compared it with previous studies, and the results show that ASR improves the throughput up to 8.6% compared with 802.11ax. ASR can avoid failure of spatial reuse, while the spatial reuse failure rate of existing schemes can up to 36%. To use the ASR scheme in practice, we investigate the relation between the optimal carrier sense threshold and transmit power. Based on the relations got from ASR, the proposed Relation based Spatial Reuse (RSR) scheme can get a satisfactory performance by using only the interference perceived and the previously found relations.