Fuzzy triangulation signature for detection of change in human emotion from face video image sequence

The present article proposes a geometry-based fuzzy relational technique for capturing gradual change in human emotion over time available from relevant face image sequences. As associated features, we make use of fuzzy membership arising out of five triangle signatures such as - (i) Fuzzy Isosceles Triangle Signature (FIS), (ii) Fuzzy Right Triangle Signature (FRS), (iii) Fuzzy Right Isosceles Triangle Signature (FIRS), (iv) Fuzzy Equilateral Triangle Signature (FES), and (v) Other Fuzzy Triangles Signature (OFS) to achieve the task of appropriate classification of facial transition from neutrality to one among the six expressions viz. anger (AN), disgust (DI), fear (FE), happiness (HA), sadness (SA) and surprise (SU). The effectiveness of the Multilayer Perceptron (MLP) classifier is tested and validated through 10 fold cross-validation method on three benchmark image sequence datasets namely Extended Cohn-Kanade (CK+), M&M Initiative (MMI), and Multimedia Understanding Group (MUG). Experimental outcomes are found to have achieved accuracy to the tune of 98.47%, 93.56%, and 99.25% on CK+, MMI, and MUG respectively vindicating the effectiveness by exhibiting the superiority of our proposed technique in comparison to other state-of-the-art methods in this regard.

     

Perfect secret sharing scheme based on vertex domination set

Due to the fast development in data communication systems and computer networks in recent years, the necessity to protect the secret data has become extremely imperative. Several methods have been proposed to protect the secret data; one of them is the secret sharing scheme. It is a method of distributing a secret K among a finite set of participants, in such a way that only predefined subset of participant is enabled to reconstruct a secret from their shares. A secret sharing scheme realizing uniform access structure described by a graph has received a considerable attention. In this scheme, each vertex represents a participant and each edge represents a minimum authorized subset. In this paper, an independent dominating set of vertices in a graph G is introduced and applied as a novel idea to construct a secret sharing scheme such that the vertices of the graph represent the participants and the dominating set of vertices in G represents the minimal authorized set. While most of the previous schemes were based on the principle of adjacent vertices, the proposed scheme is based upon the principle of non-adjacent vertices. We prove that the scheme is perfect, and the lower bound of the information rate of this new construction is improved when compared to some well-known previous constructions. We include an experiment involving security threats to demonstrate the effectiveness of the proposed scheme.     

DCT statistics-based digital dropout detection in degraded archived media

With the rapid development of visual digital media, the demand for better quality of service has increased the pressure on broadcasters to automate their error detection and restoration activities for preserving their archives. Digital dropout is one of the defects that affect archived visual materials and tends to occur in block by block basis (size of 8 × 8). It is well established that human visual system (HVS) is highly adapted to the statistics of its visual natural environment. Consequently, in this paper, we have formulated digital dropout detection as a classification problem which predicts block label based on statistical features. These statistical features are indicative of perceptual quality relevant to human visual perception, and allow pristine images to be distinguished from distorted ones. Here, the idea is to extract discriminant block statistical features based on discrete cosine transform (DCT) coefficients and determine an optimal neighborhood sampling strategy to enhance the discrimination ability of block representation. Since this spatial frame based approach is free from any motion computation dependency, it works perfectly in the presence of fast moving objects. Experiments are performed on video archives to evaluate the efficacy of the proposed technique.     

Voiced/non-voiced speech classification using adaptive thresholding with bivariate EMD

This paper introduces a robust voiced/non-voiced (VnV) speech classification method using bivariate empirical mode decomposition (bEMD). Fractional Gaussian noise (fGn) is employed as the reference signal to derive a data adaptive threshold for VnV discrimination. The analyzing speech signal and fGn are combined to generate a complex signal which is decomposed into a finite number of complex-valued intrinsic mode functions (IMFs) by using bEMD. The real and imaginary parts of the IMFs represent the IMFs of observed speech and fGn, respectively. The log-energies of both types of IMFs are calculated. There exist similarities between the IMF log-energy representation of fGn and unvoiced speech signals. Hence, the upper confidence limit from IMF log-energies of fGn is used as data adaptive threshold for VnV classification. If the subband log-energy of speech segment exceeds the threshold, the segment is classified as voiced and unvoiced otherwise. The experimental results show that the proposed algorithm performs better than the recently reported methods without requiring any training data for a wide range of SNRs.     

Alpha Fusion Adversarial Attack Analysis Using Deep Learning

The deep learning model encompasses a powerful learning ability that integrates the feature extraction, and classification method to improve accuracy. Convolutional Neural Networks (CNN) perform well in machine learning and image processing tasks like segmentation, classification, detection, identification, etc. The CNN models are still sensitive to noise and attack. The smallest change in training images as in an adversarial attack can greatly decrease the accuracy of the CNN model. This paper presents an alpha fusion attack analysis and generates defense against adversarial attacks. The proposed work is divided into three phases: firstly, an MLSTM-based CNN classification model is developed for classifying COVID-CT images. Secondly, an alpha fusion attack is generated to fool the classification model. The alpha fusion attack is tested in the last phase on a modified LSTM-based CNN (CNN-MLSTM) model and other pre-trained models. The results of CNN models show that the accuracy of these models dropped greatly after the alpha-fusion attack. The highest F1 score before the attack was achieved is 97.45 And after the attack lowest F1 score recorded is 22%. Results elucidate the performance in terms of accuracy, precision, F1 score and Recall.     

Optimal Machine Learning Enabled Performance Monitoring for Learning Management Systems

Learning Management System (LMS) is an application software that is used in automation, delivery, administration, tracking, and reporting of courses and programs in educational sector. The LMS which exploits machine learning (ML) has the ability of accessing user data and exploit it for improving the learning experience. The recently developed artificial intelligence (AI) and ML models helps to accomplish effective performance monitoring for LMS. Among the different processes involved in ML based LMS, feature selection and classification processes find beneficial. In this motivation, this study introduces Glowworm-based Feature Selection with Machine Learning Enabled Performance Monitoring (GSO-MFWELM) technique for LMS. The key objective of the proposed GSO-MFWELM technique is to effectually monitor the performance in LMS. The proposed GSO-MFWELM technique involves GSO-based feature selection technique to select the optimal features. Besides, Weighted Extreme Learning Machine (WELM) model is applied for classification process whereas the parameters involved in WELM model are optimally fine-tuned with the help of Mayfly Optimization (MFO) algorithm. The design of GSO and MFO techniques result in reduced computation complexity and improved classification performance. The presented GSO-MFWELM technique was validated for its performance against benchmark dataset and the results were inspected under several aspects. The simulation results established the supremacy of GSO-MFWELM technique over recent approaches with the maximum classification accuracy of 0.9589.     

ESPAR array design for the UHF RFID wireless sensor communication system

In this paper, in order to improve the received signal strength (RSS) and signal quality, three arrays of electronically steerable parasitic array radiator (ESPAR) antennas are suggested for the ultra-high frequency (UHF) radio frequency identification (RFID) communication and sensing system applications. Instead of the single antenna, the array antennas have recently been widely used in many communication systems because of their peak gains, better radiation patterns, and higher radiation efficiency. Also, there are some important issues to use the antenna array like high data rates in wireless communication systems and to better understand the many targets or sensors. In this article, a wireless sensor network (WSN) is being investigated to overcome multipath fading and interference by antenna nulling technology that can be achieved through beam control ESPAR array antennas. The proposed ESPAR array antennas exhibit higher gains like 9.63, 10.2, and 12 dBi and proper radiation patterns from one array to another. Moreover, we investigate the mutual coupling effect on the performance of array antennas with different spacing (0.5λ, 0.75λ, λ) and configurations. It is found that the worst mutual coupling reduced by −28 to −34 dB for 2 × 2 array, −3 to −43 dB for 2 × 3 array, and finally −42 dB to −51 dB due to the antenna spacing from 0.5λ to λ. Thus, these suggested antennas could effectively be applied in the WSN communication systems, internet of things (IoT) networks, and massive wireless and backscatter communication systems.     

A framework of genetic algorithm-based CNN on multi-access edge computing for automated detection of COVID-19

The Journal of Supercomputing - This paper designs and develops a computational intelligence-based framework using convolutional neural network (CNN) and genetic algorithm (GA) to detect COVID-19...     

Summarizing consumer reviews

Journal of Intelligent Information Systems - E-commerce giants like Amazon rely on consumer reviews to allow buyers to inform other potential buyers about a product’s pros and cons. While...     

Topological alternate centrality measure capturing drug targets in the network of MAPK pathways

A new centrality of the nodes in the network is proposed called alternate centrality, which can isolate effective drug targets in the complex signalling network. Alternate centrality metric defined over the network substructure (four nodes – motifs). The nodes involving in alternative activation in the motifs gain in metric values. Targeting high alternative centrality nodes hypothesised to be destructive free to the network due to their alternative activation mechanism. Overlapping and crosstalk among the gene products in the conserved network of MAPK pathways selected for the study. In silico knock‐out of high alternate centrality nodes causing rewiring in the network is investigated using MCF‐7 breast cancer cell line‐based data. Degree of top alternate centrality nodes lies between the degree of bridging and pagerank nodes. Node deletion of high alternate centrality on the centralities such as eccentricity, closeness, betweenness, stress, centroid and radiality causes low perturbation. The authors identified the following alternate centrality nodes ERK1, ERK2, MEKK2, MKK5, MKK4, MLK3, MLK2, MLK1, MEKK4, MEKK1, TAK1, P38alpha, ZAK, DLK, LZK, MLTKa/b and P38beta as efficient drug targets for breast cancer. Alternate centrality identifies effective drug targets and is free from intertwined biological processes and lethality.Inspec keywords: biochemistry, molecular biophysics, cellular biophysics, cancer, drugs, genetics, biomedical materialsOther keywords: MAPK pathways, complex signalling network, pagerank nodes, node deletion, drug targets, MCF‐7 breast cancer cell line‐based data, cellular mechanisms, ERK1 nodes, ERK2 nodes, MEKK2 nodes, MKK5 nodes, MKK4 nodes, MLK3 nodes, MLK2 nodes, MLK1 nodes, MEKK4 nodes, MEKK1 nodes, TAK1 nodes, P38alpha nodes, ZAK nodes, DLK nodes, LZK nodes, MLTKa/b nodes, P38beta nodes