Recognizing text in images is useful in many computer vision applications such as image search, document analysis, and robot navigation. The CR function provides an easy way to add text recognition functionality to a wide range of applications.

1. Character Recognition
Narayan Lal Menariya
Abstract:- In this Project report we have discussed
about the character Recognition(CR) process.
Recognizing text in images is useful in many
computer vision applications such as image search,
document analysis, and robot navigation. The
CR function provides an easy way to add text
recognition functionality to a wide range of
applications.
The CR functions returns the recognized text, the
recognition confidence, and the location of the text
in the original image. We can use this information
to identify the location of misclassified text within
the image.
CR performs best when the text is located on a
uniform background and is formatted like a
document. When the text appears on a non-

2. uniform background, additional pre-processing
steps are required to get the best CR results.
Purpose:- Character Recognition
In this project, we are extracting characters and numbers from
an image of a vehicle number plate.
Algorithm:-
 “Load()”-Load data from MAT-file into
workspace
S = load(filename)
 Read an image of vehicle number plate using
“imread()” command.
 Convert a color image to gray scale image and
then to an binary image using “rgb2gray()” and
“im2bw()”.
“rgb2gray()”Convert RGB image or colormap to
grayscale
I = rgb2gray(RGB)
“im2bw()”Convert image to binary image, based
on threshold

3. BW = im2bw(I, level)
Here level is taken through “graythresh”
level = graythresh(I)
 “imfill()” it is used to fill holes and thus we get
another image which do not have white spaces.
 Now remove the common part by comparing
image obtain after using “im2bw()” command
and image obtain “imfill()” command.
 Store the removed part in a variable.
 Using “bwareaopen()” command remove small
objects
BW2 = bwareaopen(BW, P)
Here BW is a variable in which removed part is
stored.
 Get the numbers of the characters using the
command “BWlabel()”
[L, num] = bwlabel(BW2)
 By using for loop, we obtained all the characters
separately.
 Resize each character according to the size of
characters present in .mat file

4.  Correlate the characters /numbers of .mat file
and characters obtained
r = corr2(A,B)
here “corr2(A,B)” computes the correlation
coefficient between A and B, where A and B are
matrices or vectors of the same size.
 Thus character recognition has been done.
3-D image
Gray Image
Binary Image Removed Part
CharactersCharacter
Recognized

5. Conclussion:-
There is some "noise" in the results due to the
smaller text next to the digits. Also, the digit 0, is
falsely recognized as the letter 'o'. This type of error
may happen when two characters have similar
shapes and there is not enough surrounding text for
the ocr function to determine the best classification
for a specific character. Despite the "noisy" results,
we can still find the digit locations in the original
image using the locateText method with the CR
results.
Another approach to improve the results is to
leverage a priori knowledge about the text within
the image.
Further processing based on a region's aspect ratio
is applied to identify regions that are likely to
contain a single character. This helps to remove the
smaller text characters that are jumbled together

6. next to the digits. In general, the larger the text the
easier it is for ocr to recognize.
Character recognition (CR) can also be used to help
the blind read normal printed text in newspapers,
novels or other books. Basically the system can
consist of the camera, OCR engine and some way of
communicating to the blind person.