Biblio

The prevalence of photo processing apps suggests the demands of picture editing. As an implementation of the convolutional neural network, style transfer has been deep investigated and there are supported materials to realize it on PC platform. However, few approaches are mentioned to deploy a style transfer model on the mobile and meet the requirements of mobile users. The traditional style transfer model takes hours to proceed, therefore, based on a Perceptual Losses algorithm [1], we created a feedforward neural network for each style and the proceeding time was reduced to a few seconds. The training data were generated from a pre-trained convolutional neural network model, VGG-19. The algorithm took thousandth time and generated similar output as the original. Furthermore, we optimized the model and deployed the model with TensorFlow Mobile library. We froze the model and adopted a bitmap to scale the inputs to 720×720 and reverted back to the original resolution. The reverting process may create some blur but it can be regarded as a feature of art. The generated images have reliable quality and the waiting time is independent of the content and pattern of input images. The main factor that influences the proceeding time is the input resolution. The average waiting time of our model on the mobile phone, HUAWEI P20 Pro, is less than 2 seconds for 720p images and around 2.8 seconds for 1080p images, which are ten times slower than that on the PC GPU, Tesla T40. The performance difference depends on the architecture of the model.

Vehicular Ad Hoc Networks (VANETs) have a strategic goal to achieve service delivery in roads and smart cities, considering the integration and communication between vehicles, sensors and fixed road-side components (routers, gateways and services). VANETs have singular characteristics such as fast mobile nodes, self-organization, distributed network and frequently changing topology. Despite the recent evolution of VANETs, security, data integrity and users privacy information are major concerns, since attacks prevention is still open issue. One of the most dangerous attacks in VANETs is the Sybil, which forges false identities in the network to disrupt compromise the communication between the network nodes. Sybil attacks affect the service delivery related to road safety, traffic congestion, multimedia entertainment and others. Thus, VANETs claim for security mechanism to prevent Sybil attacks. Within this context, this paper proposes a mechanism, called SyDVELM, to detect Sybil attacks in VANETs based on artificial intelligence techniques. The SyDVELM mechanism uses Extreme Learning Machine (ELM) with occasional features of vehicular nodes, minimizing the identification time, maximizing the detection accuracy and improving the scalability. The results suggest that the suitability of SyDVELM mechanism to mitigate Sybil attacks and to maintain the service delivery in VANETs.

Since the neural networks are utilized to extract information from an image, Gatys et al. found that they could separate the content and style of images and reconstruct them to another image which called Style Transfer. Moreover, there are many feed-forward neural networks have been suggested to speeding up the original method to make Style Transfer become practical application. However, this takes a price: these feed-forward networks are unchangeable because of their fixed parameters which mean we cannot transfer arbitrary styles but only single one in real-time. Some coordinated approaches have been offered to relieve this dilemma. Such as a style-swap layer and an adaptive normalization layer (AdaIN) and soon. Its worth mentioning that we observed that the AdaIN layer only aligns the means and variance of the content feature maps with those of the style feature maps. Our method is aimed at presenting an operational approach that enables arbitrary style transfer in real-time, reserving more statistical information by histogram matching, providing more reliable texture clarity and more humane user control. We achieve performance more cheerful than existing approaches without adding calculation, complexity. And the speed comparable to the fastest Style Transfer method. Our method provides more flexible user control and trustworthy quality and stability.

Re-drawing the image as a certain artistic style is considered to be a complicated task for computer machine. On the contrary, human can easily master the method to compose and describe the style between different images. In the past, many researchers studying on the deep neural networks had found an appropriate representation of the artistic style using perceptual loss and style reconstruction loss. In the previous works, Gatys et al. proposed an artificial system based on convolutional neural networks that creates artistic images of high perceptual quality. Whereas in terms of running speed, it was relatively time-consuming, thus it cannot apply to video style transfer. Recently, a feed-forward CNN approach has shown the potential of fast style transformation, which is an end-to-end system without hundreds of iteration while transferring. We combined the benefits of both approaches, optimized the feed-forward network and defined time loss function to make it possible to implement the style transfer on video in real time. In contrast to the past method, our method runs in real time with higher resolution while creating competitive visually pleasing and temporally consistent experimental results.

In this paper, we present an improved approach to transfer style for videos based on semantic segmentation. We segment foreground objects and background, and then apply different styles respectively. A fully convolutional neural network is used to perform semantic segmentation. We increase the reliability of the segmentation, and use the information of segmentation and the relationship between foreground objects and background to improve segmentation iteratively. We also use segmentation to improve optical flow, and apply different motion estimation methods between foreground objects and background. This improves the motion boundaries of optical flow, and solves the problems of incorrect and discontinuous segmentation caused by occlusion and shape deformation.

We investigate a deep learning model for action recognition that simultaneously extracts spatio-temporal information from a raw RGB input data. The proposed multiple spatio-temporal scales recurrent neural network (MSTRNN) model is derived by combining multiple timescale recurrent dynamics with a conventional convolutional neural network model. The architecture of the proposed model imposes both spatial and temporal constraints simultaneously on its neural activities. The constraints vary, with multiple scales in different layers. As suggested by the principle of upward and downward causation, it is assumed that the network can develop a functional hierarchy using its constraints during training. To evaluate and observe the characteristics of the proposed model, we use three human action datasets consisting of different primitive actions and different compositionality levels. The performance capabilities of the MSTRNN model on these datasets are compared with those of other representative deep learning models used in the field. The results show that the MSTRNN outperforms baseline models while using fewer parameters. The characteristics of the proposed model are observed by analyzing its internal representation properties. The analysis clarifies how the spatio-temporal constraints of the MSTRNN model aid in how it extracts critical spatio-temporal information relevant to its given tasks.

Botnets have been a major area of concern in the field of cybersecurity. There have been a lot of research works for detection of botnets. However, everyday cybercriminals are coming up with new ideas to counter the well-known detection methods. One such popular method is domain flux-based botnets in which a large number of domain names are produced using domain generation algorithm. In this paper, we have proposed a robust way of detecting DGA-based botnets using few novel features covering both syntactic and semantic viewpoints. We have used Area under ROC curve as our performance metric since it provides comprehensive information about the performance of binary classifiers at various thresholds. Results show that our approach performs significantly better than the baseline approach. Our proposed method can help in detecting established DGA bots (equipped with extensive features) as well as prospective advanced DGA bots imitating real-world domain names.

To leverage the feature extraction of DBN and the fast classification and good generalization of ELM, an improved method of DBN-ELM is proposed for intrusion detection. The improved model uses deep belief network (DBN) to train NSL-KDD dataset and feed them back to the extreme learning machine (ELM) for classification. A classifier is connected at each intermediate level of the DBN-ELM. By majority voting on the output of classifier and ELM, the final output is calculated by integration. Experiments show that the improved model increases the classification confidence and accuracy of the classifier. The model has been benchmarked on the NSL-KDD dataset, and the accuracy of the model has been improved to 97.82%, while the false alarm rate has been reduced to 1.81%. Proposed improved model has been also compared with DBN, ELM, DBN-ELM and achieves competitive accuracy.

Occurrence of faults in power systems is unavoidable but their timely recognition and location enhances the reliability and security of supply; thereby resulting in economic gain to consumers and power utility alike. Distribution Network (DN) is made smarter by the introduction of sensors and computers into the system. In this paper, detection and classification of faults in DN using Artificial Neural Network (ANN) is emphasized. This is achieved through the employment of Back Propagation Algorithm (BPA) of the Feed Forward Neural Network (FFNN) using three phase voltages and currents as inputs. The simulations were carried out using the MATLAB® 2017a. ANN with various hidden layers were analyzed and the results authenticate the effectiveness of the method.

Trademarks are recognizable images and/or words used to distinguish various products or services. They become associated with the reputation, innovation, quality, and warranty of the products. Countries around the world have offices for industrial/intellectual property (IP) registration. A new trademark image in application for registration should be distinct from all the registered trademarks. Due to the volume of trademark registration applications and the size of the databases containing existing trademarks, it is impossible for humans to make all the comparisons visually. Therefore, technological tools are essential for this task. In this work we use a pre-trained, publicly available Convolutional Neural Network (CNN) VGG19 that was trained on the ImageNet database. We adapted the VGG19 for the trademark image retrieval (TIR) task by fine tuning the network using two different databases. The VGG19v was trained with a database organized with trademark images using visual similarities, and the VGG19c was trained using trademarks organized by using conceptual similarities. The database for the VGG19v was built using trademarks downloaded from the WEB, and organized by visual similarity according to experts from the IP office. The database for the VGG19c was built using trademark images from the United States Patent and Trademarks Office and organized according to the Vienna conceptual protocol. The TIR was assessed using the normalized average rank for a test set from the METU database that has 922,926 trademark images. We computed the normalized average ranks for VGG19v, VGG19c, and for a combination of both networks. Our method achieved significantly better results on the METU database than those published previously.

In order to solve the problem that there is no effective means to find the optimal number of hidden nodes of single-hidden-layer feedforward neural network, in this paper, a method will be introduced to solve it effectively by using singular value decomposition. First, the training data need to be normalized strictly by attribute-based data normalization and sample-based data normalization. Then, the normalized data is decomposed based on the singular value decomposition, and the number of hidden nodes is determined according to main eigenvalues. The experimental results of MNIST data set and APS data set show that the feedforward neural network can attain satisfactory performance in the classification task.

SVM has been one of the most popular machine learning method for the binary classification such as sentiment analysis and spam message filtering. We explored a word embedding method for the construction of a feature vector and the deep learning method for the binary classification. CBOW is used as a word embedding technique and feedforward neural network is applied to classify SMS messages into ham or spam. The accuracy of the two classification methods of SVM and neural network are compared for the binary classification. The experimental result shows that the accuracy of deep learning method is better than the conventional machine learning method of SVM-light in the binary classification.

Electric powertrains are widely used in automotive and renewable energy industries. Reliable diagnosis for defects in the critical components such as bearings, gears and stator windings, is important to prevent failures and enhance the system reliability and power availability. Most of existing fault diagnosis methods are based on specific characteristic frequencies to single faults at constant speed operations. Once multiple faults occur in the system, such a method may not detect the faults effectively and may give false alarms. Furthermore, variable speed operations render a challenge of analysing nonstationary signals. In this work, a deep learning-based fault diagnosis method is proposed to detect common faults in the electric powertrains. The proposed method is based on pattern recognition using convolutional neural network to detect effectively not only single faults at constant speed but also multiple faults in variable speed operations. The effectiveness of the proposed method is validated via an in-house experimental setup.

Internet of Things (IoT) has an immense potential for a plethora of applications ranging from healthcare automation to defence networks and the power grid. The security of an IoT network is essentially paramount to the security of the underlying computing and communication infrastructure. However, due to constrained resources and limited computational capabilities, IoT networks are prone to various attacks. Thus, safeguarding the IoT network from adversarial attacks is of vital importance and can be realised through planning and deployment of effective security controls; one such control being an intrusion detection system. In this paper, we present a novel intrusion detection scheme for IoT networks that classifies traffic flow through the application of deep learning concepts. We adopt a newly published IoT dataset and generate generic features from the field information in packet level. We develop a feed-forward neural networks model for binary and multi-class classification including denial of service, distributed denial of service, reconnaissance and information theft attacks against IoT devices. Results obtained through the evaluation of the proposed scheme via the processed dataset illustrate a high classification accuracy.

This paper work is focused on Performance comparison of intrusion detection system between DBN Algorithm and SPELM Algorithm. Researchers have used this new algorithm SPELM to perform experiments in the area of face recognition, pedestrian detection, and for network intrusion detection in the area of cyber security. The scholar used the proposed State Preserving Extreme Learning Machine(SPELM) algorithm as machine learning classifier and compared it's performance with Deep Belief Network (DBN) algorithm using NSL KDD dataset. The NSL- KDD dataset has four lakhs of data record; out of which 40% of data were used for training purposes and 60% data used in testing purpose while calculating the performance of both the algorithms. The experiment as performed by the scholar compared the Accuracy, Precision, recall and Computational Time of existing DBN algorithm with proposed SPELM Algorithm. The findings have show better performance of SPELM; when compared its accuracy of 93.20% as against 52.8% of DBN algorithm;69.492 Precision of SPELM as against 66.836 DBN and 90.8 seconds of Computational time taken by SPELM as against 102 seconds DBN Algorithm.

Network security became a viral topic nowadays, Anomaly-based Intrusion Detection Systems [1] (IDSs) plays an indispensable role in identifying the attacks from networks and the detection rate and accuracy are said to be high. The proposed work explore this topic and solve this issue by the IDS model developed using Artificial Neural Network (ANN). This model uses Feed - Forward Neural Net algorithms and Probabilistic Neural Network and oppositional based on Particle Swarm optimization Algorithm for lessen the computational overhead and boost the performance level. The whole computing overhead produced in its execution and training are get minimized by the various optimization techniques used in these developed ANN-based IDS system. The experimental study on the developed system tested using the standard NSL-KDD dataset performs well, while compare with other intrusion detection models, built using NN, RB and OPSO algorithms.

Software insecurity is being identified as one of the leading causes of security breaches. In this paper, we revisited one of the strategies in solving software insecurity, which is the use of software quality metrics. We utilized a multilayer deep feedforward network in examining whether there is a combination of metrics that can predict the appearance of security-related bugs. We also applied the traditional machine learning algorithms such as decision tree, random forest, naïve bayes, and support vector machines and compared the results with that of the Deep Learning technique. The results have successfully demonstrated that it was possible to develop an effective predictive model to forecast software insecurity based on the software metrics and using Deep Learning. All the models generated have shown an accuracy of more than sixty percent with Deep Learning leading the list. This finding proved that utilizing Deep Learning methods and a combination of software metrics can be tapped to create a better forecasting model thereby aiding software developers in predicting security bugs.

Malware or Malicious Software, are an important threat to information technology society. Deep Neural Network has been recently achieving a great performance for the tasks of malware detection and classification. In this paper, we propose a convolutional gated recurrent neural network model that is capable of classifying malware to their respective families. The model is applied to a set of malware divided into 9 different families and that have been proposed during the Microsoft Malware Classification Challenge in 2015. The model shows an accuracy of 92.6% on the available dataset.

In this paper, we propose a deep learning framework for malware classification. There has been a huge increase in the volume of malware in recent years which poses a serious security threat to financial institutions, businesses and individuals. In order to combat the proliferation of malware, new strategies are essential to quickly identify and classify malware samples so that their behavior can be analyzed. Machine learning approaches are becoming popular for classifying malware, however, most of the existing machine learning methods for malware classification use shallow learning algorithms (e.g. SVM). Recently, Convolutional Neural Networks (CNN), a deep learning approach, have shown superior performance compared to traditional learning algorithms, especially in tasks such as image classification. Motivated by this success, we propose a CNN-based architecture to classify malware samples. We convert malware binaries to grayscale images and subsequently train a CNN for classification. Experiments on two challenging malware classification datasets, Malimg and Microsoft malware, demonstrate that our method achieves better than the state-of-the-art performance. The proposed method achieves 98.52% and 99.97% accuracy on the Malimg and Microsoft datasets respectively.

Botnet is one of the major threats on the Internet for committing cybercrimes, such as DDoS attacks, stealing sensitive information, spreading spams, etc. It is a challenging issue to detect modern botnets that are continuously improving for evading detection. In this paper, we propose a machine learning based botnet detection system that is shown to be effective in identifying P2P botnets. Our approach extracts convolutional version of effective flow-based features, and trains a classification model by using a feed-forward artificial neural network. The experimental results show that the accuracy of detection using the convolutional features is better than the ones using the traditional features. It can achieve 94.7% of detection accuracy and 2.2% of false positive rate on the known P2P botnet datasets. Furthermore, our system provides an additional confidence testing for enhancing performance of botnet detection. It further classifies the network traffic of insufficient confidence in the neural network. The experiment shows that this stage can increase the detection accuracy up to 98.6% and decrease the false positive rate up to 0.5%.

The CAPTCHA is an effective method commonly used in live interactive proofs on the Internet. The widely used CAPTCHAs are text-based schemes. In this paper, we document how we have broken such text-based scheme used by a website CAPTCHA. We use the sliding window to segment 1001 pieces of CAPTCHA to get 5900 images with single-character useful information, a total of 25 categories. In order to make the convolution neural network learn more image features, we augmented the data set to get 129924 pictures. The data set is trained and tested in AlexNet and GoogLeNet to get the accuracy of 87.45% and 98.92%, respectively. The experiment shows that the optimized network parameters can make the accuracy rate up to 92.7% in AlexNet and 98.96% in GoogLeNet.

Text-based CAPTCHAs are still commonly used to attempt to prevent automated access to web services. By displaying an image of distorted text, they attempt to create a challenge image that OCR software can not interpret correctly, but a human user can easily determine the correct response to. This work focuses on a CAPTCHA used by a popular Chinese language question-and-answer website and how resilient it is to modern machine learning methods. While the majority of text-based CAPTCHAs focus on transcription tasks, the CAPTCHA solved in this work is based on localization of inverted symbols in a distorted image. A convolutional neural network (CNN) was created to evaluate the likelihood of a region in the image belonging to an inverted character. It is used with a feature map and clustering to identify potential locations of inverted characters. Training of the CNN was performed using curriculum learning and compared to other potential training methods. The proposed method was able to determine the correct response in 95.2% of cases of a simulated CAPTCHA and 67.6% on a set of real CAPTCHAs. Potential methods to increase difficulty of the CAPTCHA and the success rate of the automated solver are considered.

Completely Automated Public Turing Test to Tell Computers and Humans Apart (CAPTCHA) is an important human-machine distinction technology for website to prevent the automatic malicious program attack. CAPTCHA recognition studies can find security breaches in CAPTCHA, improve CAPTCHA technology, it can also promote the technologies of license plate recognition and handwriting recognition. This paper proposed a method based on Convolutional Neural Network (CNN) model to identify CAPTCHA and avoid the traditional image processing technology such as location and segmentation. The adaptive learning rate is introduced to accelerate the convergence rate of the model, and the problem of over-fitting and local optimal solution has been solved. The multi task joint training model is used to improve the accuracy and generalization ability of model recognition. The experimental results show that the model has a good recognition effect on CAPTCHA with background noise and character adhesion distortion.

At present, the captcha is widely used in the Internet. The method of captcha recognition using the convolutional neural networks was introduced in this paper. It was easier to apply the convolution neural network model of simple training to segment the captcha, and the network structure was established imitating VGGNet model. and the correct rate can be reached more than 90%. For the more difficult segmentation captcha, it can be used the end-to-end thought to the captcha as a whole to training, In this way, the recognition rate of the more difficult segmentation captcha can be reached about 85%.

In practice, Defenders need a more efficient network detection approach which has the advantages of quick-responding learning capability of new network behavioural features for network intrusion detection purpose. In many applications the capability of Deep Learning techniques has been confirmed to outperform classic approaches. Accordingly, this study focused on network intrusion detection using convolutional neural networks (CNNs) based on LeNet-5 to classify the network threats. The experiment results show that the prediction accuracy of intrusion detection goes up to 99.65% with samples more than 10,000. The overall accuracy rate is 97.53%.