In this paper, a parallel deep learning model using a convolutional neural network and a dilated convolutional neural network is proposed to classify Alzheimer's disease with high accuracy in PET/CT images. The developed model consists of two pipelines, a conventional CNN pipeline, and a dilated convolution pipeline. An input image is sent through both pipelines, and at the end of both pipelines, extracted features are concatenated and used for classifying Alzheimer's disease. Complimentary abilities of both networks provide better overall accuracy than single conventional CNNs in the dataset. Moreover, instead of performing binary classification, the proposed model performs three-class classification being Alzheimer's disease, mild cognitive impairment, and normal control. Using the data received from Dong-a University, the model performs classification detecting Alzheimer's disease with an accuracy of up to 95.51%.

With the increase of motor vehicles and tourism demand, some traffic problems gradually appear, such as traffic congestion, safety accidents and insufficient allocation of traffic resources. Facing these challenges, a model of Spatio-Temporal Dilated Convolutional Network (STDGCN) is proposed for assistance of extracting highly nonlinear and complex characteristics to accurately predict the future traffic flow. In particular, we model the traffic as undirected graphs, on which graph convolutions are built to extract spatial feature informations. Furthermore, a dilated convolution is deployed into graph convolution for capturing multi-scale contextual messages. The proposed STDGCN integrates the dilated convolution into the graph convolution, which realizes the extraction of the spatial and temporal characteristics of traffic flow data, as well as features of road occupancy. To observe the performance of the proposed model, we compare with it with four rivals. We also employ four indicators for evaluation. The experimental results show STDGCN's effectiveness. The prediction accuracy is improved by 17% in comparison with the traditional prediction methods on various real-world traffic datasets.

Node identity authentication is an essential means to ensure the security of the Internet of Things. Existing blockchain-based IoT node authentication schemes have many problems. A heterogeneous IoT node authentication scheme based on an improved hybrid blockchain is proposed. Firstly, the hybrid blockchain model is designed to make the blockchain and IoT environment more compatible. Then the proxy node selection mechanism is intended to establish a bridge between the ordinary IoT node and the blockchain, building by calculating the trust value between nodes. Finally, based on the improved hybrid blockchain, the node authentication scheme of the model and proxy node selection mechanism establishes a secure connection for communication between nodes. Safety and performance analysis shows proper safety and performance.

The network capability to accomplish its functions in a timely fashion under failures and attacks is known as survivability. Ad hoc routing protocols have been studied and extended to various domains, such as Intelligent Transport Systems (ITSs), Unmanned Aerial Vehicles (UAVs), underwater acoustic networks, and Internet of Things (IoT) focusing on different aspects, such as security, QoS, energy. The existing solutions proposed in this domain incur substantial overhead and eventually become burden on the network, especially when there are fewer attacks or no attack at all. There is a need that the effectiveness of these routing protocols be analyzed in the presence of Denial of Service (DoS) attacks without any intrusion detection or prevention system. This will enable us to establish and identify the inherently stable routing protocols that are capable to survive longer in the presence of these attacks. This work presents a DoS attack case study to perform theoretical analysis of survivability on node and network level in the presence of DoS attacks. We evaluate the performance of reactive and proactive routing protocols and analyse their survivability. For experimentation, we use NS-2 simulator without detection or prevention capabilities. Results show that proactive protocols perform better in terms of throughput, overhead and packet drop.

Scheduling plays a dynamic role in cloud computing in generating as well as in efficient distribution of the resources of each task. The principle goal of scheduling is to limit resource starvation and to guarantee fairness among the parties using the resources. The demand for resources fluctuates dynamically hence the prearranging of resources is a challenging task. Many task-scheduling approaches have been used in the cloud-computing environment. Security in cloud computing environment is one of the core issue in distributed computing. We have designed a deep learning-based security model for scheduling tasks in cloud computing and it has been implemented using CloudSim 3.0 simulator written in Java and verification of the results from different perspectives, such as response time with and without security factors, makespan, cost, CPU utilization, I/O utilization, Memory utilization, and execution time is compared with Round Robin (RR) and Waited Round Robin (WRR) algorithms.

The purpose of this study is to orient users' touchpoints in co-design experience, to identify their need via visualized experience map, to recommend valid design information in online e-customization services. A user-centered co-design experience map (UCEM) is adopted to analyze the relation between users' desire and time spent, so as to evaluate the online co-design experiences. Based on evolutionary algorithm and fuzzy theory, data of this study is collected from 30 participants. The data was analyzed by descriptive analysis in SPSS, and frequency query and word cloud in NVivo. Employing design category and evaluating users' time spent, the findings are that (a) vamp color matching is consistent with interview data; (b) supported by qualitative feedback, the virtual experience map played an important role in the co-design process and the visualized interaction process; and (c) participants prefer to get more information and professional help on color matching and exterior design. Based on the findings in design category, future work should be focused on developing a better understanding of design resource recommendations and multi-stakeholder communication.

e-Governance is a medium to offer various services to citizens through a web portal, that exists in many countries nowadays. The existing e-Governance technology is a vast, centrally managed database and a set of applications that connect to it via web interfaces. Despite the modernisation of services, it remains with the lack of transparency. Thus, the existing infrastructure of e-Governance paves the way for corrupt practises by the bureaucrats. e-Governance needs a powerful underlying technology which doesn't provide any way to allow tampering of the record and which in turn eliminates corruption. In this paper, we took land registration as a use-case for building e-Governance by keeping Blockchain as an underlying technology, to put off the corrupt practices and to bring transparency. Once transactions in land registration added to the Blockchain, it is immutable as it is cryptographically secured. Besides, the blockchain technology is secured as the ledger is distributed over the network. If a hacker wants to modify the ledger, he needs to hack every node in the blockchain network. Hyperledger Fabric, a permissioned Blockchain adopted for implementation and Hyperledger Caliper for performance analysis with these evaluation metrics such as throughput, latency and execution time.

Smoke detection is helpful for early fire detection. With its large coverage area and low cost, vision-based smoke detection technology is the main research direction of outdoor smoke detection. We propose a two-stage smoke detection method combined with block Deep Normalization and Convolutional Neural Network (DNCNN) and visual change image. In the first stage, each suspected smoke region is detected from each frame of the images by using block DNCNN. According to the physical characteristics of smoke diffusion, a concept of visual change image is put forward in this paper, which is constructed by the video motion change state of the suspected smoke regions, and can describe the physical diffusion characteristics of smoke in the time and space domains. In the second stage, the Support Vector Machine (SVM) classifier is used to classify the Histogram of Oriented Gradients (HOG) features of visual change images of the suspected smoke regions, in this way to reduce the false alarm caused by the smoke-like objects such as cloud and fog. Simulation experiments are carried out on two public datasets of smoke. Results show that the accuracy and recall rate of smoke detection are high, and the false alarm rate is much lower than that of other comparison methods.

This study proposes a deep learning-based evolutionary recommendation model for heterogeneous big data integration, for which collaborative filtering and a neural-network algorithm are employed. The proposed model is used to apply an individual's importance or sensory level to formulate a recommendation using the decision-making feedback. The evolutionary recommendation model is based on the Deep Neural Network (DNN), which is useful for analyzing and evaluating the feedback data among various neural-network algorithms, and the DNN is combined with collaborative filtering. The designed model is used to extract health information from data collected by the Korea National Health and Nutrition Examination Survey, and the collaborative filtering-based recommendation model was compared with the deep learning-based evolutionary recommendation model to evaluate its performance. The RMSE is used to evaluate the performance of the proposed model. According to the comparative analysis, the accuracy of the deep learning-based evolutionary recommendation model is superior to that of the collaborative filtering-based recommendation model.

Low-rank representation methods already achieve many applications in the image reconstruction. However, for high-gradient image patches with rich texture details and strong edge information, it is difficult to find sufficient similar patches. Existing low-rank representation methods usually destroy image critical details and fail to preserve edge structure. In order to promote the performance, a new representation-based image super-resolution reconstruction method is proposed, which combines gradient domain guided image filter with the structure-constrained low-rank representation so as to enhance image details as well as reveal the intrinsic structure of an input image. Firstly, we extract the gradient domain guided filter of each atom in high resolution dictionary in order to acquire high-frequency prior information. Secondly, this prior information is taken as a structure constraint and introduced into the low-rank representation framework to develop a new model so as to maintain the edges of reconstructed image. Thirdly, the approximate optimal solution of the model is solved through alternating direction method of multipliers. After that, experiments are performed and results show that the proposed algorithm has higher performances than conventional state-of-the-art algorithms in both quantitative and qualitative aspects.

Infrared pedestrian target recognition is a vital problem of significant interest in computer vision. In this work, a novel infrared pedestrian target recognition method that uses heterogeneous features with multi-kernel transfer learning is proposed. Firstly, to exploit the characteristics of infrared pedestrian targets fully, a novel multi-scale monogenic filtering-based completed local binary pattern descriptor, referred to as MSMF-CLBP, is designed to extract the texture information, and then an improved histogram of oriented gradient-fisher vector descriptor, referred to as HOG-FV, is proposed to extract the shape information. Second, to enrich the semantic content of feature expression, these two heterogeneous features are integrated to get more complete representation for infrared pedestrian targets. Third, to overcome the defects, such as poor generalization, scarcity of tagged infrared samples, distributional and semantic deviations between the training and testing samples, of the state-of-the-art classifiers, an effective multi-kernel transfer learning classifier called MK-TrAdaBoost is designed. Experimental results show that the proposed method outperforms many state-of-the-art recognition approaches for infrared pedestrian targets.

A three-dimensional (3D) reconstruction is an important research area in computer vision. The ability to detect and match features across multiple views of a scene is a critical initial step. The tracking matrix W obtained from a 3D reconstruction can be applied to structure from motion (SFM) algorithms for 3D modeling. We often fail to generate an acceptable number of features when processing face or medical images because such images typically contain large homogeneous regions with minimal variation in intensity. In this study, we seek to locate sufficient matching points not only in general images but also in face and medical images, where it is difficult to determine the feature points. The algorithm is implemented on an adaptive threshold value, a scale invariant feature transform (SIFT), affine SIFT, speeded up robust features (SURF), and affine SURF. By applying the algorithm to face and general images and studying the geometric errors, we can achieve quasi-dense matching points that satisfy well-functioning geometric constraints. We also demonstrate a 3D reconstruction with a respectable performance by applying a column space fitting algorithm, which is an SFM algorithm.

The inspection of cracks on the surface of tunnel linings is a common method of evaluate the condition of the tunnel. In particular, determining the thickness and shape of a crack is important because it indicates the external forces applied to the tunnel and the current condition of the concrete structure. Recently, several automatic crack detection methods have been proposed to identify cracks using captured tunnel lining images. These methods apply an image-segmentation mechanism with well-annotated datasets. However, generating the ground truths requires many resources, and the small proportion of cracks in the images cause a class-imbalance problem. A weakly annotated dataset is generated to reduce resource consumption and avoid the class-imbalance problem. However, the use of the dataset results in a large number of false positives and requires post-processing for accurate crack detection. To overcome these issues, we propose a crack detection method using a ternary classifier. The proposed method significantly reduces the false positive rate, and the performance (as measured by the F1 score) is improved by 0.33 compared to previous methods. These results demonstrate the effectiveness of the proposed method.

Recently, auto-encoder has emerged as the most popular method in convolutional neural network (CNN) based image compression and has achieved impressive performance. In the traditional auto-encoder based image compression model, the encoder simply sends the features of last layer to the decoder, which cannot allocate bits over different spatial regions in an efficient way. Besides, these methods do not fully exploit the contextual information under different receptive fields for better reconstruction performance. In this paper, to solve these issues, a novel auto-encoder model is designed for image compression, which can effectively transmit the hierarchical features of the encoder to the decoder. Specifically, we first propose an adaptive bit-allocation strategy, which can adaptively select an importance channel. Then, we conduct the multiply operation on the generated importance mask and the features of the last layer in our proposed encoder to achieve efficient bit allocation. Moreover, we present an additional novel perceptual loss function for more accurate image details. Extensive experiments demonstrated that the proposed model can achieve significant superiority compared with JPEG and JPEG2000 both in both subjective and objective quality. Besides, our model shows better performance than the state-of-the-art convolutional neural network (CNN)-based image compression methods in terms of PSNR.

RSA is one of the best well-known public key cryptosystems. This methodology is widely used at present because there is not any algorithm which can break this system that has all strong parameters within polynomial time. However, it may be easily broken when at least one parameter is weak. In fact, many weak parameters are already found and are solved by some algorithms. Some examples of weak parameters consist of a small private key, a large private key, a small prime factor and a small result of the difference between two prime factors. In this paper, the new weakness of RSA is proposed. Assuming Euler's totient value, Φ (n), can be rewritten as Φ (n) = ad + b, where d is the private key and a, b ∈ ℤ, if a divides both of Φ (n) and b and the new exponent for the decryption equation is a small integer, this condition is assigned as the new weakness for breaking RSA. Firstly, the specific algorithm which is created for this weakness directly is proposed. Secondly, two equations are presented to find a, b and d. In fact, one of two equations must be implemented to find a and b at first. After that, the other equation is chosen to find d. The experimental results show that if this weakness has happened and the new exponent is small, original plaintext, m, will be recovered very fast. Furthermore, number of steps to recover d are very small when a is large. However, if a is too large, d may not be recovered because m which must be always written as m = h^a is higher than modulus.

In this paper, we propose a hash-chain based friendly force identification protocol for personal combatants equipped with a personal combat system in a tactical wireless network. It is imperative in military operations to effectively and quickly identify friendly forces. If the identification of friendly forces is not correct, this can cause friendly fire. In current ground operations, the identification of friendly forces by personal combatants is neither secure nor safe. To address this issue, the proposed protocol uses a hash-chain to determine if a detected person is friendly. Only friendly forces with the same materials that are assigned before they deploy can construct an initial hash-chain. Moreover, the hash-chain is changed at specific times. The performance of the proposed protocol is evaluated on the assumption that the secret key is leaked, which is the worst scenario in the security research field. We verify that the proposed protocol is secure for the various attack scenarios, such as message replay attack, fabrication attack, and Denial of Service attack.

The proposed framework of Four Layer Robust Storage in Cloud (FLRSC) architecture involves host server, local host and edge devices in addition to Virtual Machine Monitoring (VMM). The goal is to protect the privacy of stored data at edge devices. The computational intelligence (CI) part of our algorithm distributes blocks of data to three different layers by partially encoded and forwarded for decoding to the next layer using hash and greed Solomon algorithms. VMM monitoring uses snapshot algorithm to detect intrusion. The proposed system is compared with Tiang Wang method to validate efficiency of data transfer with security. Hence, security is proven against the indexed efficiency. It is an important study to integrate communication between local host software and nearer edge devices through different channels by verifying snapshot using lamport mechanism to ensure integrity and security at software level thereby reducing the latency. It also provides thorough knowledge and understanding about data communication at software level with VMM. The performance evaluation and feasibility study of security in FLRSC against three-layered approach is proven over 2³² blocks of data with 98% accuracy. Practical implications and contributions to the growing knowledge base are highlighted along with directions for further research.

Hybrid fuzzing which combines fuzzing and concolic execution, has proved its ability to achieve higher code coverage and therefore find more bugs. However, current hybrid fuzzers usually suffer from inefficiency and poor scalability when applied to complex, real-world program testing. We observed that the performance bottleneck is the inefficient cooperation between the fuzzer and concolic executor and the slow symbolic emulation. In this paper, we propose a novel solution named EPfuzzer to improve hybrid fuzzing. EPfuzzer implements two key ideas: 1) only the hardest-to-reach branch will be prioritized for concolic execution to avoid generating uninteresting inputs; and 2) only input bytes relevant to the target branch to be flipped will be symbolized to reduce the overhead of the symbolic emulation. With these optimizations, EPfuzzer can be efficiently targeted to the hardest-to-reach branch. We evaluated EPfuzzer with three sets of programs: five real-world applications and two popular benchmarks (LAVA-M and the Google Fuzzer Test Suite). The evaluation results showed that EPfuzzer was much more efficient and scalable than the state-of-the-art concolic execution engine (QSYM). EPfuzzer was able to find more bugs and achieve better code coverage. In addition, we discovered seven previously unknown security bugs in five real-world programs and reported them to the vendors.

An enhanced irregular repetition slotted ALOHA (IRSA) protocol is proposed by using polarization characteristic of satellite link and MIMO detection in low earth orbit (LEO) satellite networks, which is dubbed polarized MIMO IRSA (PM-IRSA). In the proposed scheme, one or two packets in one slot can be decoded by employing polarized MIMO detection, and more than two collided packets in multiple slots which can construct the virtual MIMO model can be decoded by the MIMO detection algorithm. The performance of the proposed scheme is analyzed with the density evolution (DE) approach and the degree distribution is optimized to maximize the system throughput by using a differential evolution. Numerical results certify our analysis and show that the normalized throughput of the proposed PM-IRSA can achieve 1.89 bits/symbol.

Dynamic hand gesture recognition(d-HGR) plays an important role in human-computer interaction(HCI) system. With the growth of hand-pose estimation as well as 3D depth sensors, depth, and the hand-skeleton dataset is proposed to bring much research in depth and 3D hand skeleton approaches. However, it is still a challenging problem due to the low resolution, higher complexity, and self-occlusion. In this paper, we propose a hand-shape feature extraction(HSFE) network to produce robust hand-shapes. We build a hand-shape model, and hand-skeleton based on LSTM to exploit the temporal information from hand-shape and motion changes. Fusion between two models brings the best accuracy in dynamic hand gesture (DHG) dataset.