You are on page 1of 28

St.

MARTIN’S ENGINEERING COLLEGE


UGC Autonomous
NBA & NAAC A+ ACCREDITED
Dhulapally, Secunderabad– 500100

Department of Electronics And Communication Engineering

Palm print Authentication System for Biometric Applications


Batch No: 8
1. E. Sumith(19K81A0413)
2. G. Vishal(19K81A0419)
3. G. Sai Pratheek(19K81A0420)
4. K. Rahul Reddy(19K81A0429)
Under the Guidance of
Mr. P. Pavan Kumar
Assistant professor
Department of Electronics And Communication Engineering
OUTLINE

1. Abstract
2. Introduction
3. Overall design
4. Experimental Results
5. Performance Evaluation
6. Conclusion and Future Enhancement
7. References
8. Project Code implementation
ABSTRACT

● We all have to prove our identity somewhere to access.


● So, to make personal identification more secure we use our
fingerprints, face, palm print, voice recognition and iris
● But both fingerprints and face are not so reliable in some cases.
● Therefore palm authentication has gained wide acceptance in the
networked society.

Continue…
ABSTRACT

● Therefore this project aims to develop a palm authentication system


using DWT, and textural information.
● The wavelets used for the analysis are bi-orthogonal symlet and
discrete meyer.
INTRODUCTION

• Palmprint identification had been introduced a decade ago. It is


defined as the measurement of palmprint features to recognize the
identity of a person.
• It is easy to capture using digital cameras. Palmprint does not change
much across time.

Continue…
5
INTRODUCTION

• Palm print biometric system can achieve higher accuracy than hand
geometry biometric.
• The palm print line features include principal lines, wrinkles and
ridges. Ridges are the fine lines of the palmprint.
• It requires high resolution image or inked palm print image to obtain
its features.

6
OVERALL DESIGN

● In proposed system, we use Discrete Wavelet Transform (DWT).


● DWT is used for image compression. Basically it divides image into
4x4 blocks.
● DWT has both high pass and low pass filters.
● High pass filter - extracts edges information.
● Low pass filter - extracts inner textural information.
● We also use bi-orthogonal method.
ARCHITECTURE
RESULTS OBTAINED

9
RESULTS OBTAINED

10
PERFORMANCE EVALUATION

• We as a team collectively worked together in order to meet the requirements of


the project.
• Each member of the team actively participated to make the project successful.
• We worked according to the plan we had before starting the project.
• The elements which we used in the project where under our estimated budget as
planned.
• We examined the current system of Palm print authentication system and their
drawbacks,
11
PERFORMANCE EVALUATION

• Tried to focus on the problems and worked on getting an solution to them by


this dwt based palm print authentication system.

• We noticed that there were some issues some such as the the complexity of dct
transform is way more as it divides image into 8x8 blocks.

• We studied about various transforms and their methods.

12
PERFORMANCE EVALUATION

• We divided the work among the team members equally such that one of us
worked on the database creation and other two worked on the matlab code and
other worked on the selection of quality pictures.

• Finally tried to execute the code and got the results of the project as planned.

13
CONCLUSION

• Here, we introduced a novel and highly secured biometric authentication


model with palm print identification system using morphological ROI
extraction with discrete wavelet transform(DWT).

• Due to its multi scaling functionality, two different wavelet filter banks will
be used to extract the features of distance transformed image to obtain the
most effective feature factor for comparing with a test feature vector.

• The proposed model has proven that it has achieved 100% accuracy with
several test images from the database.

14
FUTURE ENHANCEMENT

• The proposed method is highly secure and most accurate with less
time complexity for palm print authentication system.

• This method is used for high to moderate resolution images.

• In future this can be extended for low resolution images, with the
help of Artificial Intelligence and Machine Learning palm
recognition and features detection can be improved and accuracy
of the results can be improved.

15
PROJECT IMPLEMENTATION
%%
close all
warning off
%%
I = uigetfile('.jpg','select the palm image  for recognition');
I = imread(I);
if size(I,3)>1
    I = rgb2gray(I);
end
%%
level = graythresh(I);
I1 = im2bw(I,level);
K = bwmorph(I1,'dilate');
K = imrotate(K,-90);
figure,imshow(~K)
%%
D = bwdist(~K,'chessboard');
figure, imshow(D,[]), title('Distance transform of test image ');hold on;
m = size(D);
s1 = m(1)/2+100;

16
PROJECT IMPLEMENTATION
s2 = m(2)/2;
dxl = s2-128;
dxh = s2+128;
dyl = s1-128;
dyh = s1+128;
%%
p = imcrop(D,[dxl dyl dxh-100 dyh]);
figure,imshow(p,[]);title('Registered Plam Print for testing');
%%
I = imresize(I,[256 256]);
I = im2double(I);
%%
[cba1,cbh1,cbv1,cbd1] = dwt2(I,'bior3.9');
[cba2,cbh2,cbv2,cbd2] = dwt2(cba1,'bior3.9');
[cba3,cbh3,cbv3,cbd3] = dwt2(cba2,'bior3.9');
%%
[csa1,csh1,csv1,csd1] = dwt2(I,'sym8');
[csa2,csh2,csv2,csd2] = dwt2(csa1,'sym8');
[csa3,csh3,csv3,csd3] = dwt2(csa2,'sym8');
%%
[cda1,cdh1,cdv1,cdd1] = dwt2(I,'dmey');

17
PROJECT IMPLEMENTATION
[cda2,cdh2,cdv2,cdd2] = dwt2(cda1,'dmey');
[cda3,cdh3,cdv3,cdd3] = dwt2(cda2,'dmey');
%%
v1 = min(var(cbh1));
v2 = min(var(cbv1));
v3 = min(var(cbd1));
%%
v4 = min(var(cbh2));
v5 = min(var(cbv2));
v6 = min(var(cbd2));
%%
v7 = min(var(cbh3));
v8 = min(var(cbv3));
v9 = min(var(cbd3));
%%
v10 = min(var(csh1));
v11 = min(var(csv1));
v12 = min(var(csd1));
%%
v13 = min(var(csh2));
v14 = min(var(csv2));
v15 = min(var(csd2));
%%
v16 = min(var(csh3)); 18
v17 = min(var(csv3));
PROJECT IMPLEMENTATION
v18 = min(var(csd3));
%%
v19 = min(var(cdh1));
v20 = min(var(cdv1));
v21 = min(var(cdd1));
%%
v22 = min(var(cdh2));
v23 = min(var(cdv2));
v24 = min(var(cdd2));
%%
v25 = min(var(cdh3));
v26 = min(var(cdv3));
v27 = min(var(cdd3));
%%
re = [v1 v2 v3 v4 v5 v6 v7 v8 v9 v10 v11 v12 v13 v14 v15 v16 v17 v18 v19 v20 v21 v22 v23 v24 v25 v26 v27];
%%
19
FileNames = dir('test');
PROJECT IMPLEMENTATION
%%
re = [v1 v2 v3 v4 v5 v6 v7 v8 v9 v10 v11 v12 v13 v14 v15 v16 v17 v18 v19 v20 v21 v22 v23 v24 v25 v26 v27];
%%
FileNames = dir('test');
FileNames = char(FileNames.name);
[rowFN, colFN] = size(FileNames);
if(rowFN <= 2)
    FileNames = [];
    msgbox('No file in database to compare with');
else
    FileNames = [FileNames(3:end, :)];
    distmin = inf;
    k1 = 0;
    for index = 1:rowFN-2                
        eval(['load test\' FileNames(index, :) ';']);
         d = dist(re, codebook');
20
         d1 = min(d);
PROJECT IMPLEMENTATION
   dist1 = sum(min(d1,[],2)) / size(d1,1);
         if dist1 < distmin
            distmin = dist1;
            k1 = index;
            ClosestMatch = codebook;
        end      
    end
end
 dirInfo = dir('test');
%% converts the directory names to string array
dirName = char(dirInfo.name);

%% Checks if the euclidian distance is less than the threhsold only then the result would be displayed
distmin
if distmin == 0
    msg1 = sprintf(['Palm matches with ID # ' dirName(k1+2, :)]);
    msg2 = sprintf(['Palm matches with ' dirName(k1+2, :)]); 21
PROJECT IMPLEMENTATION
    k = msg1(1:28);
    f = findstr('.mat',msg1);
    f1 = msg1(f-1);
    r = strcat(k,f1);
    msgbox(r)
    fg = findstr('.mat',msg2);
    msgbox(msg2(1:fg-2));      
   else
    msg = sprintf('MATCH NOT FOUND');
    msgbox(msg);    
end
%%

22
REFERENCES

1. Laura Gulyas Oldal, Andras kovacs, “Hand geometry and palmprint based
authentication using image processing,” IEEE 18th International
Symposium on Intelligent Systems and Informatics • September, 2021 •
Subotica, Serbia

2. Vikas Varshney, Rashmi Gupta, Prema Singh, “Hybrid DWT-DCT based


Method for Palm-print Recognition,” Conference Paper · December 2021.

Continue…
REFERENCES

3. Ali Younesi and Mehdi Chehel Amirani, “Gabor Filter and Texture
based Features for Palmprint Recognition,” International Conference
on Computational Science, ICCS 2020, June 2020,Zurich, Switzerland
4. Neha Dangi and Atul Barve, “Palm Print Recognition using Steerable
Filter for Better Authentication System,” International Journal of
Computer Applications (0975 – 8887) Volume 182 – No. 15,
September 2019.

Continue…
REFERENCES

5. Ali, M.M., Yannawar .P and Gaikwad, “Study Of Edge Detection


Methods Based On Palm print lines,” International Conference on
Electrical, Electronics, and Optimization Techniques (ICEEOT) - 2019

6. G. Shobha, M. Krishna, S.C. Sharma; “ Development of Palm print


Verification System Using Biometrics “Journal of Software, Vol.17,
No.8, August 2017, pp.1824−1836.

Continue…
REFERENCES

7. Ajay Kumar, David C. M. Wong, Helen C. Shen, Anil K. Jain.


“Personal Verification using Palm print and Hand Geometry
Biometric.” Audio and Video-Based Biometric Person Authentication.
vol 2688/2003, Springer Verlag Berlin Heidelberg 2003.
8. Xinhong Zhou, Yuhua Peng, and Ming Yang. “Palm print Recognition
Using Wavelet and Support Vector Machines.” PRICAI 2006: Trends
in Artificial Intelligence, vol. 4099/2006, Springer-Verlag Berlin
Heidelberg 2006.
QUERIES ??
THANK YOU

You might also like