International Journal of Scientific & Engineering Research, Volume 6, Issue 11, November-2015 442
ISSN 2229-5518
IJSER © 2015
http://www.ijser.org
To detect text information from image,
there exist many practical difficulties, such as
non-uniform backgrounds, due to the large
variations in character font, size, texture, color,
background, orientations, and many other
reasons. Text detection from scene/text camera
images is possible due to high resolution camera.
For extracting text information from image,
algorithms are required. However extracting text
information from captured text image is difficult
due to two main factors: 1) jumbled backgrounds
with noise, text and non-text part. 2) Random text
patterns such as character, fonts, sizes etc [1].
The frequency of occurrence of text in image is
very little, and limited number of text characters
is separate from background outliers. To solve
these difficult problems captured image text is
divided into two processes: text detection and
text recognition. Text detection is used to detect
image region containing text characters. It aims to
take out non- text background outliers [3]. Text
recognition is to convert pixel based text into
readable code. Optical character recognition is
the electronic conversion of images captured by a
digital camera of printed text into readable text.
OCR has a good performance when recognizing
machine-printed text in camera-based document
analysis. Optical Character Recognition, or OCR,
is a technology that converts different types of
printed documents, such as scanned paper
documents or images captured by a digital
camera into readable data.
2. Previous Work
In this section, we present some previous
research works for assisting visually challenged
people with text to speech technology. A number
of handy reading assistants have been designed
specifically for the visually challenged [4], [5],
[6], [7], [8], [9], [10], [11], [12], [13]. Michael
R.T.F. et al. proposed a system which operates
the mobile devices without using the keypad [12].
In [14], a camera-based assistive text reading
system to read text labels and product packaging
from hand-held objects. Text detection is to
detect regions in an image that contain text
characters. [1]. Methods of feature descriptor can
broadly be classified as Histogram of the oriented
gradient (HOG) descriptor, Scale invariant
feature transform (SIFT), Speeded up robust
features (SURF), Gradient location and oriented
histogram (GLOH) [1]. These are very popular
feature descriptor used in computer vision and
image processing for the purpose of object
detection.
3. Proposed Methodology
In literature review, various text to speech
systems are discussed for visually challenged
persons but there exits some limitations. The
objective of this work is