Birla Institute of Technology and Science

About: Birla Institute of Technology and Science is a education organization based out in Pilāni, Rajasthan, India. It is known for research contribution in the topics: Computer science & Population. The organization has 8897 authors who have published 13947 publications receiving 170008 citations.

Performance Evaluation of an Adaptive-Network-Based Fuzzy Inference System Approach for Location of Faults on Transmission Lines Using Monte Carlo Simulation

Abstract: This paper employs a wavelet multiresolution analysis (MRA) along with the adaptive-network-based fuzzy inference system to overcome the difficulties associated with conventional voltage- and current-based measurements for transmission-line fault location algorithms, due to the effect of factors such as fault inception angle, fault impedance, and fault distance. This proposed approach is different from conventional algorithms that are based on deterministic computations on a well-defined model to be protected, employing wavelet transform together with intelligent computational techniques, such as the fuzzy inference system (FIS), adaptive neurofuzzy inference system (ANFIS), and artificial neural network (ANN) in order to incorporate expert evaluation so as to extract important features from wavelet MRA coefficients for obtaining coherent conclusions regarding fault location. A comparative study establishes that the ANFIS approach has superiority over ANN- and FIS-based approaches for the location of line faults. In addition, the efficacy of the ANFIS is validated through the Monte Carlo simulation for incorporating the stochastic nature of fault occurrence in practical systems. Thus, this ANFIS-based digital relay can be used as an effective tool for real-time digital relaying purposes.

Differential optical absorption spectroscopy-based refractive index sensor for cancer cell detection

Abstract: A novel spectroscopic optical sensor is presented for cancerous cell detection in various parts of the human body (i.e., cervix, adrenal gland, breast, skin, and blood). An optimized structure based on compact cladding is successfully designed with enhanced sensitivity and very low confinement loss. The parameters like effective area (Aeff), V-parameter (Veff), spot size (Weff), numerical aperture (NA), and beam quality factor are investigated over the wavelength region 1.4–2.5 µm. The consummation of relative sensitivity is also calculated and is superior to other previous work. The numerical investigation indicates that cancerous cells have higher sensitivity than normal cells. The values investigated for the sensitivity of cervical cancer, adrenal gland cancer, skin cancer, blood cancer, and breast cancer of type I and type II are 94.96%, 95.15%, 94.13%, 94.84%, 95.40%, and 95.51% in X-polarization, respectively, which are higher than the calculated values from any prior works. The explored structure is monomode PCF, with silica as a dielectric material. The finite element method (FEM) has been implemented for investigating the numerical values, implemented on COMSOL Multiphysics (version 5.3). The simple design ensures easy fabrication with ongoing techniques.

Multi-features descriptors for Human Activity Tracking and Recognition in Indoor-Outdoor Environments

Abstract: Tracking human activities and analyzing their effect in real life settings has become a task of high interest within the computer vision field as it enable many industrial and commercial applications. Sensors and communication technologies are being used for capturing human movements and providing interactive interfaces for man-machine collaboration. However, introducing intelligent computing for better scene understanding is still an unexplored domain for the researchers. In this paper, we have made an effort to allow machines understand the behaviors in outer environment by proposing a novel methodology to recognize human interactions. The objective of this research is to embed cognitive processes in information technologies for exploring new directions of intelligent media. Our proposed human activity recognition (HAR) system recognize eight complex human activities taken from BIT- Interaction dataset i.e. bow, boxing, handshake, high-five, hug, kick, pat and push. We have designed multiple features algorithms along with convolutional neural network (CNN) to evaluate the performance of our system compared with other state of the art classifiers. While, experimental results show that the proposed methodology is reliable at complex realistic settings and applicable in security systems, e-learning, smart surveillance, violence detection, child abuse protection, elderly care, virtual games, intelligent video retrievals and human computer interaction.