Improving Offline Handwritten Text Recognition
with Hybrid HMM/ANN Models
ABSTRACT:
This paper proposes the use of hybrid Hidden Markov Model (HMM)/Artificial Neural Network (ANN) models for recognizing unconstrained offline handwritten texts. The structural part of the optical models has been modeled with Markov chains, and a Multilayer Perceptron is used to estimate the emission probabilities. This paper also presents new techniques to remove slope and slant from handwritten text and to normalize the size of text images with supervised learning methods. Slope correction and size normalization are achieved by classifying local extrema of text contours with Multilayer Perceptrons. Slant is also removed in a nonuniform way by using Artificial Neural Networks. Experiments have been conducted on offline handwritten text lines from the IAM database, and the recognition rates achieved, in comparison to the ones reported in the literature, are among the best for the same task.
Existing System:
Our existing system handwritten character recognition using Modified Direction Feature (MDF), it is nothing but a system which recognize a hand written character Modified Direction Feature (MDF) generated encouraging results, reaching an accuracy of 81.58%.
In this system each and every hand written character of a separate person is scanned and stored in database the scanned images are verified using MDF.
Disadvantage of the existing system
Ø Accuracy of 81.58% is very less when compared to existing system
Ø since each and every hand written character of a separate person is scanned and stored in database it is very time consuming and it takes more manpower
Ø Since handwritten character recognition is not a most important identity of a human being this system is not widely used
Proposed system:
Our proposed system is Off-line Signature Verification using the Enhanced Modified Direction Feature and Neural-based Classification in which we are using MDF with signature images. Specifically, a number of features have been combined with MDF, to capture and investigated various structural and geometric properties of the signatures to perform verification or identification of a signature, several steps must be performed. After preprocessing all signatures from the database by converting them to portable bitmap (PBM) format, their boundaries are extracted to facilitate the extraction of features using MDF .Verification experiments are performed with classifiers We are using Radial Basis Function (RBF) which is a classifier which gives an accuracy level of 91.21%
Advantage of proposed system
Ø Accuracy level of 91.21% which very high when compared to the existing system
Ø It is very time saving
Ø It is user friendly
Hardware Requirements
• SYSTEM : Pentium IV 2.4 GHz
• HARD DISK : 40 GB
• FLOPPY DRIVE : 1.44 MB
• MONITOR : 15 VGA colour
• MOUSE : Logitech.
• RAM : 256 MB
• KEYBOARD : 110 keys enhanced.
Software Requirements
• Operating system :- Windows XP Professional
• Front End :- Microsoft Visual Studio .Net 2005
• Coding Language :- VB.NET
Modules
There are four modules available in this project they are
Ø Authentication
Ø Preprocessing
Ø Feature extraction
Ø Classification
Authentication
The first module of Off-line Signature Verification using the Enhanced Modified Direction Feature and Neural-based Classification is authentication. Authentication is done to secure the application from unauthorized user. The username and password is checked and the unauthorized user is ignored. The user can access the application if the username and password is valid. As it is the first module of the project it gives security to our application.
Preprocessing
Pre processing is nothing but a process in which input is an image the input image is converted into .pbm format which is a bitmap format And sent for further execution the purpose of converting it into bitmap format is that in second module we are going extract the boundaries of the signature if it is in bitmap format it would easy for the boundary extraction
Feature extraction
In feature extraction the boundaries of the signature image is extracted using MDF (modified extraction feature) for further modification purpose of extraction of the signatures boundaries is that
It would be easy for the classifier to identify and verify the signature because in the in the Feature extraction the size of the image is reduced.
Classification
In classification process the input is an image file the classifier verifies and identifies the signature this is the last module of this project which uses trained classifier which gives an accuracy of about 91.21% which much greater than the existing system
REFERENCE:
Salvador Espana-Boquera, Maria Jose Castro-Bleda, Jorge Gorbe-Moya and Fransisco Zamora-Martinez, “Improving Offline Handwritten Text Recognition with Hybrid HMM/ANN Models”, IEEE TRANSCATIONS ON PATTERN ANALYSIS AND MACHING INTELLIGENCE, VOL. 33, NO.4, APRIL 2011.
