Название: Computer Vision and Image Analysis for Industry 4.0
Автор: Nazmul Siddique, Mohammad Shamsul Arefin, Atiqur Rahman Ahad
Издательство: CRC Press
Год: 2023
Страниц: 214
Язык: английский
Формат: pdf (true)
Размер: 28.8 MB

Computer vision and image analysis are indispensable components of every automated environment. Modern machine vision and image analysis techniques play key roles in automation and quality assurance.

Working environments can be improved significantly if we integrate Computer Vision and image analysis techniques. The more advancement in innovation and research in Computer Vision and image processing, the greater the efficiency of machines as well as humans.

Computer Vision and Image Analysis for Industry 4.0 focuses on the roles of Computer Vision and image analysis for 4.0 IR-related technologies.
The text proposes a variety of techniques for disease detection and prediction, text recognition and signature verification, image captioning, flood level assessment, crops classifications and fabrication of smart eye-controlled wheelchairs.

A Deep Learning Approach in Detailed Fingerprint Identification: Fingerprints, as one of the most acceptable biometrics, has a significant use for security purposes. In today’s world, fingerprints are utilized in criminal investigations for identification purposes. Fingerprints can be used to identify a person’s gender, hand, finger, and other relevant aspects. The required time and effort in identifying an individual can be reduced by gender, hand, and finger classifications using fingerprints. In this chapter, a very simple model based on Convolutional Neural Networks (CNNs) is proposed to classify fingerprints by gender, hand, and finger. Convolutional Neural Network has achieved significant performance in image classification due to its ability for extraction of prominent features from complex images. CNN can extract complex features from fingerprint images and help in the process of automatic fingerprint identification and in turn, derivative characteristic details of an individual. Deriving characteristic details from a fingerprint can reduce the search space by a lot while performing automatic fingerprint detection for an individual.

Automatic Image Captioning Using Deep Learning: Transforming an image to a text or descriptive form has recently gained a lot of research appeal. Creating a sentence with correct semantics and syntactic structure is still a matter of concern. Object recognition, relations between the objects, and different meanings of the same word make this task more difficult. Therefore, an inspection of the attention mechanism has recently achieved great progress. In this chapter, we describe some existing models and then expand these models by integrating BERT, LSTM, and dense models together. After analysing the results, we found that while training on the same parameters, our new model has shown comparatively less training time than the others and has shown better results for all common metrics (BLEU, METEOR, & CIDEr) on the MS-COCO dataset. Image captioning is the technique of interpreting a source image into its corresponding text version. The captioning system helps the audience rapidly grasp the image’s concept without having to go over every element. The ultimate goal of an image captioning system is to communicate the picture’s primary message in a narrative format. Image captioning is a category of image-to-sequence problem in which the sources are pixels (a digital form of an image).