I. Introduction
Textual images are images that have text in them. These textual pictures can either be scanned or taken with a camera. Scanned textual pictures are captured with a scanner, whereas camera caught textual images are obtained with a digital camera or a smartphone camera. Scanners can be used to capture printed document and scene text images, while digital cameras and cell phone cameras can capture printed document and scene text images. Digital cameras are a powerful, efficient, and widely used image capture technology found on virtually all portable devices, including phones, tablets, pens, watches, and PDAs [1]. Due to technical improvements, the market for Document Imaging Analysis is changing. Sales of built- in camera devices such as phones and tablets have risen sharply, while scanner sales have declined. This is because these cameras are used in a variety of applications. So, to find the text that has been noticed. The backdrop is gradually isolated from the set text in the remaining part. According to the [2] different studies the use of the terms text verification, detection, and localization has been used interchangeably.
Schematic representation of scene text detection and recognition. This image example is from total text (Chng and chan, 2017) [7].