Description

Signature Verification through MATLAB Using
Image Processing
Kritika Raghuwanshi
Niketa Dubey,
Riju Nema
Rishabh Sharma
kritikaraghuwanshi.17@gmail.com nikita112dubey@gmail.com
riju.nema27@gmail.com
rish1990786@gmail.com
(B.E), EC Deptt. Jai Narain College Of Technology & Science, Bhopal(India)
Abstract: The signature is a particular pattern
used by human beings for their personal
identification. To prevent the forgery many
software are used now a days. Signature
verification system can be

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.

Related Documents

Share

Transcript

Available online @ www.ijetecs.com (pp:203-206), © 2013 ijetecs
203
Signature Verification through MATLAB Using Image Processing
Kritika Raghuwanshi Niketa Dubey, Riju Nema Rishabh Sharma
kritikaraghuwanshi.17@gmail.com nikita112dubey@gmail.com riju.nema27@gmail.com rish1990786@gmail.com
(B.E), EC Deptt. Jai Narain College Of Technology & Science, Bhopal(India)
Abstract: The signature is a particular pattern used by human beings for their personal identification. To prevent the forgery many software are used now a days. Signature verification system can be of two types, either offline or online signature verification system. In this paper we verify the offline signature, which is acquired through a camera. This signature verification system is designed using MATLAB
.
Keywords:
Data Acquisition, Image acquisition, Image Processing, Off-line signature, Gray-scale
.
I. Introduction
Biometric methods are nowadays being increasingly used to for identification purposes both due to its accuracy and convenience at the user/consumer end. There exist a number of biometrics methods today e.g. Signatures, Fingerprints, Iris etc. There is considerable interest in authentication based on handwritten signature verification system as it is the cheapest way to authenticate the person. a signature is treated as an image carrying a certain pattern of pixels that pertains to a specific individual. Signature Verification Problem therefore is concerned with determining whether a particular signature truly belongs to a person or not. There are two approaches to signature verification, online and offline differentiated by the way data is acquired. In offline case signature is obtained on a piece of paper and later scanned. We are recognizing and verifying the signatures through digital image processing using MATLAB software so that we get the exact results with high precision and much feasible for diverse applications. We approach the problem in two steps. Initially the scanned signature image is preprocessed to be suitable for extracting features. Then the preprocessed image is used to extract relevant geometric parameters that can distinguish signatures of different persons. The net step involves the use of these extracted features to verify a given images.
II. Literature Survey
Signature Data Base For training and testing of the signature recognition and verification system 10 signatures are used. The templates of the signature as shown in figure 1. The input signature is captured from the digital high pixel camera. The next stage was the preprocessing of these signatures collected. Figure.1 Signature Templates The following were the processes that were carried out on the signatures.
Conversion from RGB image to a black and white (logical) image- For signature verification, the color of ink has no significance at all. Instead the form of two signatures must be compared. Hence all scanned images were converted to black and white images where white is represented by 1 and black by 0. Hence the signature part of the image was represented by 0 and blank part of the image (without any signature) by 1. This conversion also makes future coding easier.
Available online @ www.ijetecs.com (pp:203-206), © 2013 ijetecs
204
Using region props to filter out those regions having area less than 45 pixels- The region props function of MATLAB was used to remove all stray dots arising in the image due to improper scanning (noise). By setting the threshold for area at 45 pixels, all arbitrary dots get removed, but it is ensured that the actual signature is not affected as its area is usually in the order of 1000s of pixels.
Finding the bounding box of the image and cropping unnecessary parts (edges) of the image- The bounding box of the image was found and the image was cropped out, thereby eliminating blank area from the sides of the image. Thus now each image consists only of the signature part. We have avoided resizing the images to the same width or height as it distorts the image to a large extent. The above steps have been graphically shown on figure 2. Figure.2 Conversion from RGB to Gray scale.
III. Feature Extraction
For building of signature verification system, we have used VPP and HPP information VPP (vertical projection profile)is nothing but the sum of all pixels at each x coordinate, plotted versus x. On the other hand HPP (horizontal projection profile) is the sum of all pixels at each y coordinate, plotted versus y.
It must be kept in mind that after the pre-processing described before, each pixel containing a signature has the logical value of 1.Thus, for a particular x coordinate, whenever the values of all pixels along y are summed up, it gives an estimate of the intensity of the signature at that x coordinate .Thus, by similar reasoning, we can conclude that the VPP and HPP profiles qualitatively describe the distribution of the signature along the horizontal (x) and vertical (y) axis respectively. Now we elaborate on how the feature vector for VPP (and HPP) is found. First, a signature, say s1, is taken and its VPP (or HPP) vector found. Then it is compared against the VPP (or HPP) vectors of all other specimen signatures (s2 through s10).For each comparison (e.g. between s1 and s2), Firstly the two VPP (or HPP) profiles are resized using the imresize function of MATLAB so that both profiles are of the same length. The intensity values of s1 and s2 are subtracted, and the absolute value of the difference is stored in another vector called holder. An example of a signature along with its VPP and HPP plots is shown as. The VPP profile of the above signature is shown below. It can be observed that the VPP profile takes the maximum value at around x=130, where the letter
‘p’ occurs in the actual signature.
The HPP profile of the same signature is shown below.
Available online @ www.ijetecs.com (pp:203-206), © 2013 ijetecs
205
IV. MATHEMATICAL ANALYSIS
The camera gives a colored image at the MATLAB. The colored image is of the RGB (Red, Green, Blue) format. In other terms, the colored image consist of coordinate axis and three color matrices i.e. Red, Green and Blue color matrices. But the process which we are carrying out in
MATLAB is based on Gray scale format. Therefore we convert our color image into Gray scale through equation no. 1 Gray color = 0.299*Red + 0.5876*Green + 0.114*Blue
…
(1) The VPP and HPP profiles qualitatively describe the distribution of the signature along the horizontal (x) and vertical (y) axis respectively. And they can mathematically be derived from equation no. 2 and 3 VPP (x) =
… (2)
HPP (y) =
… (3)
Where A (x,y) represents the logical value (either 0 or 1) at the pixel with coordinate (x,y), and xmax is the maximum x coordinate of the signature and ymax is maximum y coordinate of the signature. After the images have been matched in the MATLAB the present error is given in terms of two parameters i.e. Root Mean Square (RMS) error and Signal to Noise Ratio (SNR). These can be calculated mathematically from equation 4 and 5. MeanErr = MeanErr + sum(sum(SqrErr(:,:,i)))/(row*col);
… (4)
PSNR= 10*log10(c*255^2/MeanErr); … (5)
V. CONCLUSION
In this study, we presented Off-Line Signature Recognition and Verification System through MATLAB using image processing. While evaluating the signature verification system we evaluated two parameters as detailed before
–
VPP (Vertical Projection Profile) and HPP (Horizontal Projection Profile). From testing performed we noticed that the testing done on the basis of VPP is more reliable than testing done on the basis of HPP. VPP provides a more reliable result on tests like false acceptance, false rejection and failure result rather than HPP.
Generally, the threshold is set to obtain the best possible overall result, taking into account all the tests, as tweaking the threshold is invariably a trade off between the accuracy for various tests (viz false acceptance, false rejection, and forgeries). However, the success rate for detection of forgeries is quite low. This is because while signing a forgery, the person makes a conscious attempt to make the forged signature as similar to the srcinal signature as possible. In general, for all the three grounds of evaluation, the performance improves if the size of the database, i.e. the number of comparisons is increased. This study aims to reduce to a minimum the cases of forgery in business transactions.
REFERENCES [1] Anu Rathi, Divya Rathi, Parmanand Astya
Offline handwritten Signature Verification by using Pixel based Method
International Journal of Engineering Research & Technology (IJERT) ISSN: 2278-0181 Vol. 1 Issue 7, September-2012. [2] Nilesh Y. Choudhary
,
Mrs. Rupal Patil, Dr. Umesh. Bhadade and
Prof. Bhupendra M Chaudhari
Signature Recognition & Verification System Using Back Propagation Neural Network
International Journal of IT, Engineering and Applied Sciences Research (IJIEASR) ISSN: 2319-4413 Volume 2, January 2013.
Available online @ www.ijetecs.com (pp:203-206), © 2013 ijetecs
206
[3] Prashanth C. R. and K. B. Raja
Off-line Signature Verification Based on Angular Features
International Journal of Modeling and Optimization, Vol. 2, No. 4, August 2012. [4] Aniketh Talwai, B.S. Sasikanth and Debabrata
Sengupta “
Offline Signature Verification” Project
Work Report, IIT Guwahati, April 2009.
Authors Profile: 1.
KRITIKA RAGHUWANSHI Is
Under Graduate in Electronics and Communications Engineering from the Rajiv Gandhi Technical University, from Jai Narain College of Technology and Science, Bhopal, M.P., India. 2013. 2.
NIKETA DUBEY Is
Under Graduate in Electronics and Communications Engineering from the Rajiv Gandhi Technical University, from Jai Narain College of Technology and Science, Bhopal, M.P., India. 2013. 3.
RIJU NEMA
Is Under Graduate in Electronics and Communications Engineering from the Rajiv Gandhi Technical University,. from Jai Narain College of Technology and Science, Bhopal, M.P., India. 2013 4.
RISHABH SHARMA
Is Under Graduate
in Electronics and Communications Engineering from the Rajiv Gandhi Technical University, from Jai Narain College of Technology and Science, Bhopal, M.P., India. 2013.

We Need Your Support

Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks