Blockchain and Deep Learning for Smart Healthcare

12Computer Vision Techniques for Smart Healthcare Infrastructure

Reshu Agarwal

Amity Institute of Information Technology, Amity University, Noida, India

Abstract

This Chapter discusses the use of computer vision for optical character recognition (OCR) of cursive writing; that is, the aim was to identify cursive style texts that were destroyed due to motion blur while in a slow-moving vehicle. The field of application is transportation, one of the many fields where the use of machine learning for OCR has been found extremely useful. OCR helps identify destroyed vehicle nameplates, read messages and texts displayed through printed means such as billboards and visual electronic means such as TV, and read traffic signs and boards, which are just some of its wide range of applications. When we click a picture of a billboard or any printed poster from a moving vehicle, various disturbances, particularly the motion blur effect, influence the image acquired, which makes it hard to read the text or make out the image. The OCR carried out by machines can easily read and predict such non-readable texts. Modern research studies have proposed powerful techniques such as deep learning, machine learning, binary format matrix techniques, and Internet of Things (IoT), among others. The idea behind this project was highly influenced by such techniques, which uses a line detection method combined with a circular empty mask to detect and identify the edge pixels of targeted texts, crops the selected ...