Journey of scene text components recognition: Progress and open issues
- PDF / 1,964,155 Bytes
- 26 Pages / 439.37 x 666.142 pts Page_size
- 73 Downloads / 214 Views
Journey of scene text components recognition: Progress and open issues Payel Sengupta 1 & Ayatullah Faruk Mollah 1 Received: 17 December 2019 / Revised: 27 August 2020 / Accepted: 9 September 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
In computer vision, scene text component recognition is an important problem in end-toend scene text reading systems. It involves two major sub-problems - segmentation of such components into scene characters and classification of segmented characters into known character classes. Significant attention and increasingly focused research efforts are being put forth and reasonable progress in this field has already been made, though a diversity of challenges like background complexity, variety of text appearances, noise, blur, distortion and various other degradation and deformation issues are still left to address. In this paper, we present (i) a detail survey of scene component segmentation and/or recognition methods reported so far in literature, (ii) related datasets available for quantitative evaluation and benchmarking segmentation and/or recognition performance, (iii) comparative results and analysis over the reported methods, and (iv) discussion on open areas to be looked into in order to achieve the desired goal of end-to-end scene text recognition. Moreover, this paper provides an acceptable reference for researcher in the area of scene text components segmentation and recognition. Keywords Scene text components . Scene character segmentation . Scene character recognition . Datasets of scene components
1 Introduction Nowadays, as a product of information enabled human civilization, digital information in the form of scene images are available in plenty due to easy availability of camera phones and internet communication. This necessitates an important task of extracting meaningful information like contact number, names, locations, date, time, etc. from such images. To retrieve these substantial
* Ayatullah Faruk Mollah [email protected]
1
Department of Computer Science and Engineering, Aliah University, IIA/27 New Town, Kolkata 700160, India
Multimedia Tools and Applications
information appropriately is one of the enormous challenges in computer vision. It involves two important sub-problems, viz. (i) localization of text regions from an image, and (ii) recognition of text present therein. While the first problem has received significant attention over last two decades and considerable progress has been made, the latter is yet to receive such attention. Therefore, segmentation and/or recognition of localized text components, also referred to as scene text components, is an essential task in scene image analysis. A scene text component may have multiple text lines, a line may have multiple words and words usually have multiple characters. Line and word segmentation may be considered as a trivial problem in comparison to character segmentation and subsequent recognition. Moreover, many text localization methods attempt to locali
Data Loading...
