Abstract: Writing is the
painting of the voice and handwriting enables civilization. All of us have different handwritings.
It is difficult to recognize the different kind of handwritings, especially the
doctor’s prescription. Often the same medicine are prescribed for different
kinds of diseases. The aim of this paper is to propose a system which uses
curvelet transform and artificial neural network for the recognition of
doctor’s prescription and convert it into a record. Then retrieve the content of
the medicine. The input is a scanned image of prescription and output is a bill
with the prescription and the content of medicine. The Curvelet transform is to
be used in the feature extraction stage and artificial neural network is used
for recognizing prescription. Curvelet transform makes it easier to extract
curves in handwriting. Back propagation algorithm is employed to train the
system. So this system helps us to know whether the prescribed medicine is
right. It also gives a solution to the difficulties in understanding a
OCR, ANN, Feature extraction and
Every individual has different kind of handwriting.
Some handwritings are beautiful and some are not. It is easy for human beings
to read and understand a handwritten document. But a system cannot recognize
the different kind of handwriting. By using the OCR is able to provide that
ability to the system. It is easy for human beings to read and understand a
handwritten document. By using the OCR is able to provide that ability to the
system. The conversion of hand written image or text in to document or record
is called optical character recognition (OCR). Handwriting recognition refers
to understanding or determining the written word and converting it into a
printed format. This technology is using different fields including banking,
postal, teaching etc.
is classified into two types. They are handwritten character recognition and
printed character recognition. Hand written character is again divided in to
two on-line and of-line character recognition. There are several advantages for
OCR. It can reduce the data entry time. It can reduce the storage space
required by the time. The other advantage is fast retrieval of the data.
are many recognition systems to recognize the English handwritten document.
This paper focus on the recognition of a prescription it is very tough to
understand the matter in it. The prescription will be written in cursive
writing. It also will have many curves in it.
So by using curvelet transform we can easily extract the features of
character. Artificial neural network (ANN) is used for the classification. ANN
is a computational model. The aim of ANN is to provide the human intelligence
to the machines. After classifying the character the system aims to retrieve
the content of the medicine.
2. Related works
There exist many systems for the
classifying handwritten characters. Most of the systems do not support the
cursive letters. A handwritten recognition system must have 2 steps. They are feature
extraction and classification. The systems mainly make use of wavelet transform
for the feature extraction. Different algorithms from neural network are used
for the classification. In some system support vector machine is used for the
classification. Many researchers have developed
the character recognition systems by using template matching, spatial features,
Fourier and shape descriptors, Normalized chain code, Invariant moments, central
moments, Zernike moments, modified invariantmoments,structural,statistical,Topological,Gabor,Zoning
features combinations of these features etc. Different pattern classifiers like
neural networks, Hidden Markov models, and Fuzzy and SVM classifiers are used.
paper we propose a system to recognize the medical prescription and retrieve
the content of the medicine. The system includes 5 modules. The first four modules
are for the recognition of the prescription. The last step is to retrieve the
content of the medicine. Modules for classifying the prescription include
preprocessing, segmentation, feature extraction, and classification.
Classification is done using artificial neural network. A neural network is
trained with the 26 characters of English language. The features of the character
which is to be recognized are given as input to the system. The neural network
compares the input features with the trained data set in it. After classifying
or recognizing the letter it returns the letter. After classifying the entire
medicine it is given as an input to medical database. Medical database returns
t1e information on content of the medicine. The modules implemented in this
paper are shown in the fig 1.The proposed system architecture is shown in fig
Collection of sample data for training the neural
network is involved in this module. Data from different sources are collected
and stored in a file. The recognition system acquires a scanned image as
an input image. The image should have a specific format such as JPEG, BMT etc.
There will be many irregularities in the scanned
prescription due to the sporadic handwriting. So the scanned image cannot give
directly to the system as input. The irregularities affect the performance of
recognition system badly. So some operations should be performed on the image
to remove their irregularities and to make them in a normalized form.
Preprocessing is done to remove this kind of irregularities in order to get a
better performance. Preprocessing includes three functionalities. They are
Firstly the cropping of images were done manually.
Then the size of all images is made as uniform. Then the noises form the image
is removed by using median filtering algorithm. Secondly the process of
binarization is done which makes our image as a binary image. It is done by
global thresholding method. Now the image is reduced to level intensities white
and black. After inverting the image the boundary box is created for every word
which touch the four sides of the word. At last thinning is
done to resize the image.
Image segmentation is a process of separating the
image in the super pixels. Segmentation makes the image more meaningful. It is
easy to analyses a segmented image. The scanned prescription contains the names
of medicine. The name is separated in to a single character for further
proceedings. The individual character is obtained by the character
Feature extraction is used to reduce the
dimensionality of the image.it is done to extract the unique features or
property of every single character in the prescription. By extracting the
unique features we can define a letter with minimum amount of resources. The
letter can be represented with lesser number of bits. Curvelet transform is
used for the feature extraction because prescription contains many curves in
it. More focus is made on Discrete
Curvelet Transform with the Wrapping Technique.
Algorithm for Feature Extraction
Input: image after segmentation
Output: features library
segmented image of 64X64 pixels
image is reduced by using a discrete curvelet transform with a wrapping
find out the curvelet coefficient for every characters
4: compute the standard deviation
of these coefficients in order to get a feature set of input
5: obtain the features of every
single character in the image and store it on a train library.
Classification refers to the recognition of the
character.it is done by using a multi-layer perceptron. Neural network is used
for recognition. Before applying neural network it has to be trained with
character database. The input to the trained neural network is the features of
the character that is to be recognized. Neural network is already trained with
26 characters and its features.it compares the input with this data and return
the most matched pattern as the result. The neural network classifies the input
into one of the 26 characters.
Algorithm for classification
Input: Isolated test character images.
Output: recognition of prescription
1. Obtain the features as per the algorithm.
2. Store these feature vectors in test library database.
3. Compute the % of similarity between the features in the test library and
4. Obtain the character with maximum % of similarity and print that character.
Obtaining the details of the medicine
After recognizing the letter next step is to
retrieve the content of medicine. For this a medical database is created. The
recognized medicine is given as input to the medical database. It compares with
medicine and the content of medicine to the user.
An algorithm proposed here is used for the
recognition of medical prescription .The system is expected to give a high
performance with the maximum accuracy. Curvelet transform is used for the feature
extraction. It will be easier because the prescription contains many curves in
handwriting. ANN is used to provide the artificial intelligence to the system.
Back propagation algorithm is used to classify the prescription.at last the
text document of the prescription obtained as an output with the content of
medicine in it. This system helps to solve the dilemma in understanding the prescription.
We hereby express our sincere thanks to our
dear teachers and other staffs for
their inestimable and overwhelming support. We would like to express deep sense of gratitude to our guide Ms Anitha L, Asst. professor of
department of computer science and engineering for her encouragement and
guidance for the successful completion of this paper.
We would also like to express our heartfelt thanks to
our beloved parents and friends for their blessings and moral support.
1 Sandeep Saha, Sayam Kumar, Nabarag
Paul,.Sandip Kundu. “Optical Character recognition
using 40-point feature extraction and artificial neural network”,International
Journal of Advanced Research in Computer Science and Software Engineering 2013 APRIL 4.
2 Shanjana c, Ajay,” Offline Recognition of Malayalam Handwritten Text” 8TH international conference inter discplinarity
in engineering,2014 October.
, Prof.I.Muthumani ,” Optical
Character Recognition for Handwritten Cursive English characters”,International
Journal of Computer Science and Information Technologies.
Parameshwarppa.and V. Dhandra “Handwritten
Kannada Characters Recognition using Curvelet Transform”, International
Journal of Computer Applications, National conference on Digital Image and
Signal Processing, DISP 2015.
5 R. Abd. Rahim1, L.U.Abdul Khalik, M.N.S.Zainudin
“Handwritten English Character Recognition Using Gradient Feature Extraction”,
international journal for advance research in enginrreing and technology
(IJARET). Volume 3, Issue XII, Dec. 2015.
Mangesh A, Navale Ganesh D , Sharma Hemant P, Nikam Krushna V, Shaikh Riyaj R,
“HCR(English) using Neural Network”,
IJARIIE-ISSN(O) Vol-1 Issue-4 2015.
Manuel, Associate Professor, and Saidas S.R. ” Handwritten Malayalam Character Recognition using Curvelet Transform and
ANN”, International Journal of Computer Applications (IJCA).Volume 121 –
No.6, July 2015.