Machine Vision Inspection Systems, Machine Learning-Based Approaches

Muthukumaran Malarvel, Soumya Ranjan Nayak, Prasant Kumar Pattnaik, Surya Narayan Panda

• Publisher: Wiley-Scrivener
• Year: 2021
• Print ISBN: 9781119786092
• eBook ISBN: 9781119786108
• Edition: 1
• Topic: Computer Science
• Subtopic: Artificial Intelligence (AI) & Semantics
• Index: Computer Science

Machine Vision Inspection Systems (MVIS) is a multidisciplinary research field that emphasizes image processing, machine vision and, pattern recognition for industrial applications. Inspection techniques are generally used in destructive and non-destructive evaluation industry. Now a day's the current research on machine inspection gained more popularity among various researchers, because the manual assessment of the inspection may fail and turn into false assessment due to a large number of examining while inspection process.

This volume 2 covers machine learning-based approaches in MVIS applications and it can be employed to a wide diversity of problems particularly in Non-Destructive testing (NDT), presence/absence detection, defect/fault detection (weld, textile, tiles, wood, etc.), automated vision test & measurement, pattern matching, optical character recognition & verification (OCR/OCV), natural language processing, medical diagnosis, etc. This edited book is designed to address various aspects of recent methodologies, concepts, and research plan out to the readers for giving more depth insights for perusing research on machine vision using machine learning-based approaches.

1
Machine Learning-Based Virus Type Classification Using Transmission Electron Microscopy Virus Images

Kalyan Kumar Jena^1*, Sourav Kumar Bhoi¹, Soumya Ranjan Nayak² and Chittaranjan Mallick³

¹Department of Computer Science and Engineering, Parala Maharaja Engineering College, Berhampur, India

²Amity School of Engineering and Technology, Amity University Uttar Pradesh, Noida, India

³Department of Mathematics, Parala Maharaja Engineering College, Berhampur, India

Abstract

Viruses are the submicroscopic infectious agents having the capability of replication itself inside the living cells of human body. Different dangerous infectious viruses greatly affect the human society along with plants, animals and microorganisms. It is very difficult for the survival of human society due to these viruses. In this chapter, Machine Learning (ML)-based approach is used to analyze several transmission electron microscopy virus images (TEMVIs). In this work, several TEMVIs such as Ebola virus (EV), Entero virus (ENV), Lassa virus (LV), severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), Zika virus (ZV), etc. are analyzed. The ML-based approach mainly focuses on the classification techniques such as Logistic Regression (LR), Neural Network (NN), k-Nearest Neighbors (kNN) and Naive Bayes (NB) for the processing of TEMVIs. The performance of these techniques is analyzed using classification accuracy (CA) parameter. The simulation of this work is carried out using Orange3-3.24.1.

Keywords: ML, TEMVIs, Classification Techniques, LR, NN, kNN, NB

1.1 Introduction

ML [1–34] plays an important role in the today’s era for the researchers and scientists to carry out their research work. ML is considered as one of the most important application of artificial intelligence. Systems can be learned and improved from experience in automatic manner without any explicit programming by using ML mechanism. The main focus of ML is to develop computer programs that can access data as well as use it for learning purpose. ML techniques can be mainly classified as unsupervised learning techniques and supervised learning techniques. Unsupervised learning techniques focus on clustering techniques and supervised learning techniques focus on classification techniques. Hierarchical clustering, distance map, distance matrix, DBSCAN, manifold learning, k-means, Louvain clustering, etc. are some ML-based clustering techniques. ML [1–34] focuses on several classification techniques such as LR, NN, kNN, NB, decision tree, random forest, AdaBoost, etc. The similar objects can be grouped into a set which is known as cluster by using clustering techniques. Classification techniques are used to categorize a set of data into classes. In classification technique, the algorithm can learn from the data input provided to it and then use this learning mechanism to classify new observations. These techniques are mainly used to categorize the data into a desired and distinct number of classes where label can be assigned to each class. It is a very challenging task to categorize the set of data into classes accurately. Several ML-based classification techniques can be used for such classification. Viruses [57, 58] are the submicroscopic infectious agents and they are having the replication capability due to which they replicate itself inside the living cells of human body. Viruses can be classified as DNA and RNA viruses on the basis of nucleic acid, cubical, spiral, and radial symmetry, complex viruses on the basis of structure, bacteriophage, plant and animal, insect viruses on the basis of host range. Several viruses can be transmitted through respiratory route, feco-oral route, sexual contacts, blood transfusion, etc. Very dangerous viruses such as SARS-CoV-2, EV, ENV, LV, ZV, dengue virus, Hepatitis C virus have adverse effects which greatly affect the human society in the current scenario. In this work, several ML-based classification techniques such as LR, NN, kNN, NB are focused for the implementation of classification mechanism on several TEMVIs such as EV, ENV, LV, SARS-CoV-2 and ZV.

The main contribution of this work is stated as follows.

• ML-based approach is used for the processing of several TEMVIs ...

Table of contents

• Cover
• Table of Contents
• Title Page
• Copyright
• Preface
• 1 Machine Learning-Based Virus Type Classification Using Transmission Electron Microscopy Virus Images
• 2 Capsule Networks for Character Recognition in Low Resource Languages
• 3 An Innovative Extended Method of Optical Pattern Recognition for Medical Images With Firm Accuracy—4f System-Based Medical Optical Pattern Recognition
• 4 Brain Tumor Diagnostic System— A Deep Learning Application
• 5 Machine Learning for Optical Character Recognition System
• 6 Surface Defect Detection Using SVM-Based Machine Vision System with Optimized Feature
• 7 Computational Linguistics-Based Tamil Character Recognition System for Text to Speech Conversion
• 8 A Comparative Study of Different Classifiers to Propose a GONN for Breast Cancer Detection
• 9 Mexican Sign-Language Static-Alphabet Recognition Using 3D Affine Invariants
• 10 Performance of Stepped Bar Plate-Coated Nanolayer of a Box Solar Cooker Control Based on Adaptive Tree Traversal Energy and OSELM
• 11 Applications to Radiography and Thermography for Inspection
• 12 Prediction and Classification of Breast Cancer Using Discriminative Learning Models and Techniques
• 13 Compressed Medical Image Retrieval Using Data Mining and Optimized Recurrent Neural Network Techniques
• 14 A Novel Discrete Firefly Algorithm for Constrained Multi-Objective Software Reliability Assessment of Digital Relay
• Index
• End User License Agreement