Preface

摘要

The present era of technological advancements is moving at a very fast pace. It is therefore essential to keep oneself abreast with all the recent developments in various engineering domains. The book discusses various state-of-the-art developments in the diverse area of communications, data processing and signal processing, and the endeavor has been to bring together some of these developments in a concise platform that would benefit students, researchers, academicians and industry people. The chapters presented in the book have been selected on the basis of relevance and mathematical deliberations on the topics. Apart from the above-listed domains, this book has additionally included topics on social issues providing advanced technological solutions.Chapter "Deep Semantic Segmentation for Self-driving Cars" introduces the technique of semantic segmentation of urban scene for a self-driving car that comprises three sub-systems in navigation, viz. lane finding, urban scene understanding and geo-positioning. In Chapter "Shot Boundary Detection Using Artificial Neural Network," hybrid video shot boundary detection process using feature extraction by mean log difference is discussed in combination with artificial neural network techniques. A system for leaf parameter analysis is proposed in Chapter "Custard Apple Leaf Parameter Analysis, Leaf Diseases, and Nutritional Deficiencies Detection Using Machine Learning," where detection of N, P and K deficiencies and leaf diseases is accomplished using K-nearest neighbors (k-NN) and support vector machine (SVM) algorithms. A typical problem of recognizing and removing the rain streaks on photographs by an improved convolutional neural network (CNN) architecture is discussed in Chapter "Single Image Rain Removal Using Convolutional Neural Network." A study of voice samples for two disorders- hypo and hyper-along with normal voice samples is considered in Chapter "A Robust Approach of Estimating Voice Disorder Due to Thyroid Disease" to create a databank for three classes-normal, hypo and hyper. A combined classifier, i.e., SVM and HMM (hidden Markov model), was utilized.