This paper solves some face recognition problems including segmenting, extracting, and identifying facial features that are thought to face from the background. The optimization introduced, allowed to decrease in the number of scales from 46 to 7 and the number of processed sub-windows from 8,443,722 to 704 which implies a reduction of 99.99% of processed sub-windows. Also, it is possible to coexist the two parallel programming frameworks utilized in our application (Fast Flow and OpenMP) to exploit all the available resources on CPU Multi-Core Platforms. ✻ Index: Web of Science, Scopus (Q3), Inspec, Ei Compendex, Semantic Scholar, EBSCOhost

Our work represents the whole procedure including road image acquisition, pre-processing the image, detecting the road, and measurement of the road from various camera distances. This research work ended up with an accuracy of 96% and we have tried to capture the images with a 90-degree view. We used is quite suitable for a cost-effective solution.

This work covers future frame prediction and proposes a recurrent network model which utilizes recent techniques from deep learning research. The presented architecture is based on the recurrent decoderencoder framework with convolutional cells, which allows the preservation of Spatio temporal data correlations. Driven by perceptual-motivated objective functions and a modern recurrent learning strategy, it can outperform existing approaches concerning future frame generation in several video content types with the MNIST dataset, MsPacman dataset, and UCF-101 dataset. All this can be achieved with fewer training iterations and model parameters.

The main objective of this research is to gain high accuracy using as little CPU Time as possible, keeping into consideration the facts like lighting conditions, vehicle motion, noisy plates, and segmented words in the input image. Also, extract a clean image of the license plates of private or community vehicles. We have gained an overall accuracy of 98.41% in plate detection, 82.25% in plate extraction, and 94.11% in character segmentation, resulting in 76.19% overall accuracy using the available dataset. That solve real-life problem such as processing the input image up to Character Segmentation so that it is easy to recognize the character by sending it to any existing Optical Character Recognition (OCR) system. ✻ Index: SCOPUS, zbMATH, SCImago

We advocate using CNN in conjunction with basic statistical characteristics that keep info regarding the global shape of different time series for local feature extraction. We also look at how the duration of a time series affects recognition accuracy with the WISDM dataset, restricting it to a single second to allow for time series analysis activity classification. ✻ Index: SCOPUS, INSPEC, WTI Frankfurt eG, zbMATH, SCImago

We define the DDoS attacks detection method based on artificial intelligence and explored with more than 96 percent accuracy a technique to detect a DDoS attack assault danger using artificial intelligence (AI). In addition to a secure or healthy network, authors have identified 7 separate sub-categories (BENIGN, UDPLag, Syn, UDP, MSSQL, Portmap, NetBIOS) of DDoS attacks. ✻ Index: DBLP, EI Compendex, INSPEC, SCImago, SCOPUS, zbMATH

We used the CICIDS2017 dataset to meet the objective and propose an IDS based on the deep learning technique to increase safety. The purpose was to use a neural network classifier to predict the network and web attacks (Brute Force, Web Attack XSS, Web Attack SQL Injection, DoS Hulk, DoS GoldenEye, Heartbleed, FTPPatator, SSHPatator, DoS Slowloris, DoS Slowhttptest). This research mainly aims to model a neural network classifier that can predict 14 network/web attacks and regular traffic with 91% accuracy. ✻ Index: ISI Conference Proceedings Citation Index - ISI Web of Science, Google Scholar, Scopus, DBLP

We used the Aegean Wi-Fi Intrusion Dataset (AWID3), 1.8 million tuples of reduced data. This paper used the reduced version and predicted if an attack was one of four types using the k-nearest-neighbors classifier. The primary goal of the paper is to attain the highest accuracy possible when creating a model that is capable of classifying the 4 attack types (MOK, Deauthentication, Authentication Request, and ARP) and detecting and classifying wireless attacks using a Machine Learning models dataset. Another goal supported by this main objective was determining a way to avoid the curse of dimensionality. Our best results were for the attack “arp” type where we attained the best accuracy with recall. ✻ Index: SCOPUS, INSPEC, WTI Frankfurt eG, zbMATH, SCImago

We provided a mechanism to improve confidentiality in communications by splitting packet flows (each flow) into different physical paths. This makes snooping attacks less effective. The document will consider the provisioning of multipath TCP through multipath routes. In practice, multipath communication requires making IP packets follow different routes, which is not possible at the network level using standard Internet routing. This paper aims at providing complete solutions to address private communication from the client to the cloud, i.e., to transfer data over channels without exposing its content.

In this paper, we study several reinforcement learning algorithms, ranging from asynchronous qlearning to deep reinforcement learning. We focus on the improvements they provide over standard rei nforcement learning algorithms, as well as the impact of initial starting conditions on the performance of a reinforcement learning agent ✻ Index: SCOPUS, INSPEC, WTI Frankfurt eG, zbMATH, SCImago

This research paper summarizes the evolution of the big data concept and the techniques that are closely connected to the big data concept, including the CAP theorem, which happens to be one of the most important rules in modern distributed computing. Moreover, this paper focuses on open-source software built from the Apache big data platform. It looks at the domain of the NoSQL database Cassandra, which is the primary software considered in this study, and the real use cases of Cassandra and related technologies in real-world scenarios. The proposed study also describes the development of certain use cases.

We have proposed a smart contract-based secure data-sharing scheme in healthcare by leveraging the advantages of the interplanetary file system (IPFS). This thesis is to attain higher levels of medical records security using Ethereum Blockchain methods. Finally, it also helps organ donation, to incentivize medical stakeholders such as researchers, health authorities, etc. to participate in the network as blockchain miners. This provides them with access to aggregate, anonymous data mining awards in return for sustaining, and securing the network.