Blood Group Identification Based on Fingerprint by Using 2D Discrete Wavelet and Binary Transform

Blood Group Identification Based on Fingerprint by Using 2D Discrete Wavelet and Binary Transform

M. Mondal U.K. Suma M. Katun R. Biswas* Md. Rafiqul Islam

Mathematics Discipline, Science, Engineering and Technology School, Khulna University, Khulna 9208, Bangladesh

Department of Mathematics, Bangladesh University, Dhaka 1207, Bangladesh

Corresponding Author Email: 
rajibkumath11@gmail.com
Page: 
57-70
|
DOI: 
https://doi.org/10.18280/mmc_c.802-404
Received: 
1 April 2019
|
Accepted: 
22 May 2019
|
Published: 
30 December 2019
| Citation

OPEN ACCESS

Abstract: 

Fingerprint is becoming the part of our day to day life right from our home to workplace. Now a day for security and safety purpose prime importance is given by it. Also, Fingerprint identification is one of the most popular biometric technologies and which is highly used in criminal investigations, commercial applications, and so on. The performance of a fingerprint image-matching algorithm depends heavily on the quality of the input fingerprint images. It is very important to acquire good quality images to find out gender and blood group. The use of wavelet transform improves the quality of an image and reduces noise level. In this research, different compression techniques are used to overcome this problem.  Initially, different finger prints have been collected among the students of Mathematics discipline of Khulna University. Then the fingerprints have been resized to 320*256 and saved in JPEG format. Then, we have used different wavelets transformation for compression of the fingerprint images. Image qualities before compression and after compression are measured by Mean Squared Error (MSE), Signal-to-Noise Ratio (SNR) and Peak Signal-to-Noise Ratio (PSNR). After that the images has been pre-processing and transformed into binary images. Further, from this binary image, pixel calculation such that numbers of black and white pixels have been calculated. After that, some mathematical calculations have been done to identify the genders and blood groups. This work has been done by MATLAB programming.

Keywords: 

discrete wavelets, binary transform, blood group recognition, finger print

1. Introduction

Now a day, Fingerprint is most welcome objects to our human life. In criminal investigation, fingerprint concerns a vital role because every fingerprint has an individual characteristic there was no same pattern found between two persons. Also, fingerprint pattern will remain unchanged for the life of an individual; however, the print itself may change due to permanent scars and skin diseases. Further fingerprints have general characteristic ridge patterns that allow them to be systematically identified. Fingerprint is a made of a series of ridges and valleys on the surface of the finger. Due to this characteristic of fingerprint, this is very essential to justify any criminal. Also, fingerprint is used in gender and blood group identification.

Today, fingerprint recognition has been around the longest, and there are more commercial applications of it than iris recognition. For example, different types of fingerprint recognition devices can be found for network access and physical access entry configurations. A Study of fingerprint patterns in relation to gender and blood group has been presented by Narayana et al. [1]. A study of fingerprints in relation to gender and blood group was studied by Rastogi et al. [2]. Pattern of fingerprints in different ABO blood groups was published by Bharadwaja et al. [3]. Also, finger print identification was discussed by Surinder [4]. Further, relationship between pattern of fingerprints and blood groups was submitted by Smail et al. [5]. These types of work are done by different researcher with different observation as [6-16].

This research work is to be researched the blood group identification based on fingerprint by using 2D discrete wavelet and binary transform. Initially, different fingerprints have been collected, resized and saved in JPEG format. Then from this RGB image, we have found out its binary image and its black and white pixels number. Also, the length of binary number of black pixels, number of 0’s and number of 1’s are calculated. All have been done by MATLAB programming. After finding these values, we have used 4 steps of calculations. Then these 4 steps of calculations, which is identifies the different blood groups of the given fingerprints.

2. Experimental Data with All Information's

Table A: Experimental data with all information such as, serial number of fingerprints, name, original fingerprints, Gray scale fingerprints, Black and white fingerprints, gender and blood group has been shown in the following Table A:

Person ID

Original Fingerprints

Gray Scale Fingerprints

Black and White Fingerprints

Gender

Blood Group

1

Male

B+

2

Male

AB+

3

Female

O+

4

Male

AB+

5

Male

B+

6

Male

O+

7

Male

A+

8

Female

B+

9

Female

O+

10

Female

B+

11

Male

B+

12

Male

AB+

13

Female

O+

14

Male

B+

15

Female

O+

16

Male

O+

17

Male

O+

18

Male

AB+

19

Female

A+

20

Male

AB+

21

Male

AB+

22

Male

B+

23

Female

AB+

24

Male

B+

25

Male

B+

26

Male

O+

27

Female

A+

28

Female

AB+

29

Male

B+

30

Male

O+

31

Male

O+

32

Female

AB+

33

Male

B+

34

Male

A+

35

Male

AB+

36

Female

A+

37

Female

A+

38

Male

B+

39

Female

A+

40

Female

A+

41

Female

AB+

42

Female

O+

43

Female

A+

44

Female

O+

45

Male

A+

46

Male

AB+

47

Female

B+

48

Male

O+

49

Male

A+

50

Male

AB+

51

Female

A+

52

Female

B+

53

Male

B+

54

Female

A+

55

Female

O+

56

Female

O+

57

Female

AB+

58

Female

AB+

59

Female

A+

60

Female

A+

3. Working Procedure

In this experiment, we have collected 60 fingerprints (30 male fingerprints and 30 female fingerprints) with their only Rh+ ABO blood groups. For blood group identification, we have separated the 4 Rh+ ABO groups like, A+, B+, O+ and AB+. After performed these 4 steps calculations, we have obtained a good result where the following procedure has been done:

(1) At first, we have separated the all fingerprints into 4 Rh+ ABO groups (A+, B+, O+ and AB+) from the obtained prepossessing fingerprints.

(2) Number of White Pixels (NWP) and Number of Black Pixels (NBP) and Binary Length of Black Pixels (BLBP) for each fingerprint has been taken, which is already find out in chapter 3. In this chapter, we have taken those values to do perform the calculations.

(3) Subtraction (NBP-NWP) from Number of Black Pixels (NBP) to Number of White Pixels (NWP), number of 0’s (total number of digits 0) and 1’s (total number of digits 1) in the Binary Length of Black Pixels (BLBP) has been find out due to proper calculations of blood identification.

(4) All calculations have been done by MATLAB programming.

(5) The following steps has been followed for blood group identification:

Step-1

For identification of blood group A+,

Calculate Quotient $=\frac{\text { NBP }}{0^{\text {'}}sof \text { BLBP }}$ using MATLAB programming. Also, odd number quotient has been noted among the four blood groups.

Step-2

For identification of blood group B+,

Calculate Quotient $=\frac{\text { NWP-NBP }}{\text { B LBP }}$ using MATLAB programming. Also, odd number quotient has been noted among the four blood groups.

Step-3

For identification of blood group AB+,

Calculate Modulo $=\frac{\text { NBP }}{1^{\text {'}} \text {sof BLBP }}$ using MATLAB programming. Also, even number modulo has been noted among the four blood groups.

Step-4

For identification of blood group O+,

Calculate Quotient $=\frac{\text { NWP }}{1^{\text {'}} \text {sof BLBP }}$ using MATLAB programming. Also, even number quotient has been noted among the four blood groups.

4. Experimental Results

Experimental results are shown in the following table and discussions have been mention in the end of the Table 1.

Here, in Table 2. Number of Even Quotient =03 and Number of Odd Quotient=12.

So, the percentage of Number of Odd Quotient=80%.

Table 1. Results for step-1 on 15 fingerprints for blood group A+

Person ID

NWP

NBP

BLBP

0’s

Quotient

Observation

7

13700

68220

17

11

6201

Odd

19

71431

10483

14

06

1747

Odd

27

70526

11394

14

10

1139

Odd

34

13915

68005

17

10

6800

Even

36

77195

4725

13

06

787

Odd

37

81077

843

10

04

211

Odd

39

69828

12092

14

06

2015

Odd

40

67257

14663

14

06

2443

Odd

43

71680

10240

14

12

853

Odd

45

38205

43715

16

07

6245

Odd

49

22838

59082

16

07

8440

Even

51

69865

12055

14

05

2411

Odd

54

69590

12330

14

08

1541

Odd

59

73640

8280

14

09

920

Even

60

78790

3130

12

06

521

Odd

 
Table 2. Results for step-1 on 15 fingerprints for blood group B+

Person ID

NWP

NBP

BLBP

0’s

Quotient

Observation

1

6849

75071

17

07

10274

Even

5

31095

50825

16

09

5647

Odd

8

78270

3650

12

07

521

Odd

10

78076

3844

12

07

549

Odd

11

32525

49395

16

08

6174

Even

14

28542

53378

16

11

4852

Even

22

9327

72593

17

09

8065

Odd

24

17486

64434

16

05

12886

Even

25

63069

18851

14

06

3141

Odd

29

41857

40063

16

05

8012

Even

33

26842

55078

16

09

6119

Odd

38

19872

62048

16

09

6894

Even

47

29735

52185

13

06

8698

Even

52

67893

14027

14

05

2804

Even

53

16695

65225

16

05

13044

Even

 
Here, Number of Even Quotient =09 and Number of Odd Quotient=06.

So, the percentage of Number of Odd Quotient=40%.

Table 3. Results for step-1 on 15 fingerprints for blood group AB+

Person ID

NWP

NBP

BLBP

0’s

Quotient

Observation

 2

31004

50916

16

08

6364

Even

4

26925

54995

16

06

9165

Odd

12

24927

56993

16

07

8141

Odd

18

18560

63360

16

08

7920

Even

20

23744

58176

16

10

5817

Odd

21

15086

66834

17

13

5141

Odd

23

80744

1176

11

07

168

Even

28

78256

3664

12

07

523

Odd

32

68948

12972

14

07

1853

Odd

35

31352

50568

16

10

5056

Even

41

76483

5437

13

05

1087

Odd

46

29735

52185

16

06

8698

Even

50

32549

49371

16

08

6172

Even

57

67596

14324

14

04

3582

Even

58

72630

9290

14

06

1548

Even

 
Here, Number of Even Quotient =08 and Number of Odd Quotient=7.

So, the percentage of Number of Odd Quotient=46.67%.

Table 4. Results for step-1 on 15 fingerprints for blood group O+

Person ID

NWP

NBP

BLBP

0’s

Quotient

Observation

3

27129

54791

16

08

6848

Even

6

9922

71998

17

09

7999

Even

9

67555

14365

14

07

2123

Odd

13

71480

10440

14

09

1160

Even

15

80744

1176

11

07

168

Even

16

12796

69124

17

12

5760

Even

17

36663

45257

16

09

5028

Even

26

3676

78244

17

10

7824

Even

30

18596

63324

16

05

12664

Even

31

32955

48965

16

06

8160

Even

42

79563

2357

12

06

392

Even

44

79810

2010

11

05

402

Even

48

31004

50916

16

08

6364

Even

55

75560

6360

13

07

908

Even

56

67993

13927

14

05

2785

Odd

 
Here, Number of Even Quotient =13 and Number of Odd Quotient=02.

So, the percentage of Number of Odd Quotient=13.33%.

Table 5. Results for step-2 on 15 fingerprints for blood group A+

Person ID

NWP

NBP

NWP-NBP

BLBP

Quotient

Observation

7

13693

68227

54534

17

3207

Odd

19

71431

10483

60948

14

4353

Odd

27

70526

11394

59132

14

4223

Odd

34

13915

68005

54090

17

3181

Odd

36

77195

4725

72470

13

5574

Even

37

81077

843

80234

10

8023

Odd

39

69828

12092

57736

14

4109

Odd

40

67257

14663

52594

14

3756

Even

43

71680

10240

61440

14

4338

Even

45

38205

43715

5510

16

344

Even

49

22838

59082

36244

16

2265

Odd

51

69865

12055

57810

14

4129

Odd

54

69590

12330

57260

14

4090

Even

59

73640

8280

65360

14

4668

Even

60

78790

3130

75660

12

6304

Even

 
Here, Number of Even Quotient =07 and Number of Odd Quotient=08.

So, the percentage of Number of Odd Quotient=53.33%.

Table 6. Results for step-2 on 15 fingerprints for blood group B+

Person ID

NWP

NBP

NWP-NBP

BLBP

Quotient

Observation

1

6849

75071

68222

17

4013

Odd

5

31095

50825

19730

16

1233

Odd

8

78270

3650

74620

12

6218

Even

10

78076

3844

74232

12

6187

Odd

11

32525

49395

16870

16

1053

Odd

14

28542

53378

24836

16

1551

Odd

22

9327

72593

63266

17

3721

Odd

24

17486

64434

46948

16

2933

Odd

25

63069

18851

44218

14

3158

Even

29

41857

40063

1794

16

113

Odd

33

26842

55078

28236

16

1763

Odd

38

19872

62048

42176

16

2635

Odd

47

29735

52185

22450

13

1725

Odd

52

67893

14027

53866

14

3847

Odd

53

16695

65225

48530

16

3033

Odd

 
Here, Number of Even Quotient =02 and Number of Odd Quotient=13.

So, the percentage of Number of Odd Quotient=86.66%.

Table 7. Results for step-2 on 15 fingerprints for blood group AB+

Person ID

NWP

NBP

NWP-NBP

BLBP

Quotient

Observation

 2

31004

50916

19912

16

1244

Even

4

26925

54995

28070

16

1754

Even

12

24927

56993

32066

16

2004

Even

18

18560

63360

44800

16

2800

Even

20

23744

58176

34432

16

2152

Even

21

15086

66834

51748

17

3044

Even

23

80744

1176

79568

11

7233

Odd

28

78256

3664

74592

12

6216

Even

32

68948

12972

55976

14

3998

Even

35

31352

50568

19216

16

1201

Odd

41

76483

5437

71046

13

5465

Odd

46

29735

52185

22450

16

1403

Odd

50

32549

49371

16822

16

1051

Odd

57

67596

14324

53272

14

3805

Odd

58

72630

9290

63340

14

4524

Even

 
Here, Number of Even Quotient =10 and Number of Odd Quotient=06.

So, the percentage of Number of Odd Quotient=40%.

Table 8. Results for step-2 on 15 fingerprints for blood group O+

Person ID

NWP

NBP

NWP-NBP

BLBP

Quotient

Observation

3

27129

54791

27662

16

1778

Even

6

9922

71998

62076

17

3651

Odd

9

67555

14365

53190

14

3799

Odd

13

71480

10440

61040

14

4360

Even

15

80744

1176

79568

11

7233

Odd

16

12796

69124

56328

17

3313

Odd

17

36663

45257

8594

16

537

Odd

26

3676

78244

74568

17

4336

Even

30

18596

63324

44728

16

2795

Odd

31

32955

48965

16010

16

1000

Even

42

79563

2357

77206

12

6433

Odd

44

79810

2010

77800

11

7072

Even

48

31004

50916

19912

16

1244

Even

55

75560

6360

69200

13

5323

Odd

56

67993

13927

54066

14

3861

Odd

 
Here, Number of Even Quotient =06 and Number of Odd Quotient=09.

So, the percentage of Odd Number of Quotient=60%.

Table 9. Results for step-3 on 15 fingerprints for blood group A+

Person ID

NWP

NBP

BLBP

1’s

Modulo

Observation

7

13693

68227

17

06

1

Odd

19

71431

10483

14

08

3

Odd

27

70526

11394

14

05

4

Even

34

13915

68005

17

07

0

Even

36

77195

4725

13

07

0

Even

37

81077

843

10

06

3

Odd

39

69828

12092

14

09

5

Odd

40

67257

14663

14

08

7

Odd

43

71680

10240

14

02

0

Even

45

38205

43715

16

06

5

Odd

49

22838

59082

16

09

6

Even

51

69865

12055

14

09

4

Even

54

69590

12330

14

05

0

Even

59

73640

8280

14

05

0

Even

60

78790

3130

12

06

4

Even

 
Here, Number of Even Modulo =09 and Number of Odd Modulo=06.

So, the percentage of Number of Even Modulo =60%.

Table 10. Results for step-3 on 15 fingerprints for blood group B+

Person ID

NWP

NBP

BLBP

1’s

Modulo

Observation

1

6849

75071

17

10

1

Odd

5

31095

50825

16

07

5

Odd

8

78270

3650

12

05

0

Even

10

78076

3844

12

05

4

Even

11

32525

49395

16

08

3

Odd

14

28542

53378

16

05

3

Odd

22

9327

72593

17

08

1

Odd

24

17486

64434

16

11

3

Odd

25

63069

18851

14

10

1

Odd

29

41857

40063

16

11

1

Odd

33

26842

55078

16

09

7

Odd

38

19872

62048

16

07

0

Even

47

29735

52185

13

07

0

Even

52

67893

14027

14

09

5

Odd

53

16695

65225

16

11

4

Even

 
Here, Number of Even Modulo =05 and Number of Odd Modulo=10.

So, the percentage of Number of Even Modulo =33.33%.

Table 11. Results for step-3 on 15 fingerprints for blood group AB+

Person ID

NWP

NBP

BLBP

1’s

Modulo

Observation

 2

31004

50916

16

08

4

Even

4

26925

54995

16

10

5

Odd

12

24927

56993

16

09

5

Odd

18

18560

63360

16

08

0

Even

20

23744

58176

16

06

0

Even

21

15086

66834

17

04

2

Even

23

80744

1176

11

04

0

Even

28

78256

3664

12

05

4

Even

32

68948

12972

14

07

1

Odd

35

31352

50568

16

06

0

Even

41

76483

5437

13

08

5

Odd

46

29735

52185

16

10

5

Odd

50

32549

49371

16

08

4

Even

57

67596

14324

14

10

6

Even

58

72630

9290

14

08

2

Even

 
Here, Number of Even Modulo =10 and Number of Odd Modulo=05.

So, the percentage of Number of Even Modulo =66.67%.

Table 12. Results for step-3 on 15 fingerprints for blood group O+

Person ID

NWP

NBP

BLBP

1’s

Modulo

Observation

3

27129

54791

16

08

5

Odd

6

9922

71998

17

08

6

Even

9

67555

14365

14

07

1

Odd

13

71480

10440

14

05

0

Even

15

80744

1176

11

04

0

Even

16

12796

69124

17

05

4

Even

17

36663

45257

16

07

3

Odd

26

3676

78244

17

07

5

Odd

30

18596

63324

16

11

8

Even

31

32955

48965

16

10

5

Odd

42

79563

2357

12

06

5

Odd

44

79810

2010

11

06

0

Even

48

31004

50916

16

08

4

Even

55

75560

6360

13

06

0

Even

56

67993

13927

14

09

4

Even

 
Here, Number of Even Modulo =09 and Number of Odd Modulo=06.

So, the percentage of Number of Even Modulo =60%.

Table 13. Results for step-4 on 15 fingerprints for blood group A+

Person ID

NWP

NBP

BLBP

1’s

Quotient

Observation

7

13693

68227

17

06

2282

Even

19

71431

10483

14

08

8928

Even

27

70526

11394

14

05

14105

Odd

34

13915

68005

17

07

1987

Odd

36

77195

4725

13

07

11027

Odd

37

81077

843

10

06

11582

Even

39

69828

12092

14

09

7758

Even

40

67257

14663

14

08

8407

Odd

43

71680

10240

14

02

35840

Even

45

38205

43715

16

06

6367

Odd

49

22838

59082

16

09

2537

Odd

51

69865

12055

14

09

7762

Even

54

69590

12330

14

05

13918

Even

59

73640

8280

14

05

14728

Even

60

78790

3130

12

06

13131

Odd

 
Here, Number of Even Quotient =08 and Number of Odd Quotient=07.

So, the percentage of Number of Even Quotient=53.33%.

Table 14. Results for step-4 on 15 fingerprints for blood group B+

Person ID

NWP

NBP

BLBP

1’s

Quotient

Observation

1

6849

75071

17

10

685

Odd

5

31095

50825

16

07

4442

Even

8

78270

3650

12

05

15654

Even

10

78076

3844

12

05

15615

Odd

11

32525

49395

16

08

4065

Odd

14

28542

53378

16

05

5708

Even

22

9327

72593

17

08

1165

Odd

24

17486

64434

16

11

1589

Odd

25

63069

18851

14

10

6306

Even

29

41857

40063

16

11

3805

Odd

33

26842

55078

16

09

2982

Even

38

19872

62048

16

07

2838

Even

47

29735

52185

13

07

4247

Odd

52

67893

14027

14

09

7543

Odd

53

16695

65225

16

11

1517

Odd

 
Here, Number of Even Quotient =06 and Number of Odd Quotient=09.

So, the percentage of Number of Even Quotient=40%.

Table 15. Results for step-4 on 15 fingerprints for blood group AB+

Person ID

NWP

NBP

BLBP

1’s

Quotient

Observation

 2

31004

50916

16

08

4250

Even

4

26925

54995

16

10

2692

Even

12

24927

56993

16

09

2769

Odd

18

18560

63360

16

08

2320

Even

20

23744

58176

16

06

3957

Odd

21

15086

66834

17

04

3771

Odd

23

80744

1176

11

04

20186

Even

28

78256

3664

12

05

15651

Odd

32

68948

12972

14

07

9849

Odd

35

31352

50568

16

06

5225

Odd

41

76483

5437

13

08

9560

Even

46

29735

52185

16

10

2973

Odd

50.

32549

49371

16

08

4068

Even

57

68570

13350

14

10

6857

Odd

58

72630

9290

14

08

9078

Even

 
Here, Number of Even Quotient =07 and Number of Odd Quotient=08.

So, the percentage of Number of Even Quotient=46.67%.

Table 16. Results for step-4 on 15 fingerprints for blood group O+

Person ID

NWP

NBP

BLBP

1’s

Quotient

Observation

3

27129

54791

16

08

3392

Even

6

9922

71998

17

08

1240

Even

9

67555

14365

14

07

9650

Even

13

71480

10440

14

05

14296

Even

15

80744

1176

11

04

20176

Even

16

12796

69124

17

05

2569

Odd

17

36663

45257

16

07

5238

Even

26

3676

78244

17

07

525

Odd

30

18596

63324

16

11

1690

Even

31

32955

48965

16

10

5495

Odd

42

79563

2357

12

06

13302

Even

44

79810

2010

11

06

13260

Even

48

31004

50916

16

08

3876

Even

55

75560

6360

13

06

12592

Even

56

67993

13927

14

09

7554

Even

 
Here, Number of Even Quotient =12 and Number of Odd Quotient=03.

So, the percentage of Number of Even Quotient=80%.

5. Results and Discussion

From the above Table-1, Table-2, Table-3 and Table-4, we can see that the percentage of Odd Number Quotient for blood group (A+) is greater than others blood groups. So, we conclude that, when we apply Step-1 for any one fingerprint whose blood group is unknown, his/her blood group probability (A+) will be 80%.

Also, we can see that the percentage of Odd Number Quotient for blood group (B+) is greater than others blood groups. So, we conclude that, when we apply Step-2 for any one fingerprint whose blood group is unknown, his/her blood group probability (B+) will be 86.67%. This observation is shown in the Table-5 to Table-8.

Further, the percentage of Even Number of Modulo for blood group (AB+) is greater than others blood groups. So, we conclude that, when we apply Step-3 for any one fingerprint whose blood group is unknown, his/her blood group probability (AB+) will be 66.67% all have are presented in the Table-9 to Table-12.

Again, from the above Table-13 toTable-16, we can see that the percentage Number of Even Quotient for blood group O+ is greater than others blood groups. So, we conclude that, when we apply Step-4 for any one fingerprint whose blood group is unknown, his/her O+ blood group probability will be 80%.

Therefore we conclude that, when we apply Step-1 for any one fingerprint whose blood group is unknown, his/her A+ blood group probability will be 80%; for Step-2 probability of B+ blood group will be 80.67%; for Step-3 probability of AB+ blood group will be 66.67% and for Step-4 probability of O+ blood group will be 80% which are shown in the above Table-1 to Table-16.

  References

[1] Narayana, B.L., Rangaiah, Y.K.C., Khalid, M.A. (2016). Study of fingerprint patterns in relation to gender and blood group. J Evolution Med. Dent. Society, 5(14): 530-533. https://doi.org/10.14260/jemds/2016/144

[2] Rastogi, D.P., Ms Pillai, K.R. (2010). A study of fingerprints in relation to gender and blood group. J Indian Acad Forensic Med, 32(1): 11-14. 

[3] Bharadwaja, A., Saraswat, P.K., Agrawal, S.K. (2004). Pattern of fingerprints in different ABO blood groups. Journal of Forensic Medicine and Toxicology, 21(2): 49-52.

[4] Surinder, N. (1984). Finger Print Identification. Gita Press, 1-15. https://doi.org/10.1016/S0140-6736(00)51333-6

[5] Smail, H.O., Wahab, D.A., Abdullah, Z.Y. (2019). Relationship between pattern of fingerprints and blood groups. Journal of Advanced Laboratory Research in Biology, 10: 84-90.

[6] Agrawal, H., Choubey, S. (2014). Fingerprint based gender classification using multi-class SVM. International Journal of Advanced Research in Computer Engineering & Technology, 3(8): 2575-2580.

[7] Arulkumaran, T., Sankaranarayanan, D.P.E., Sundari, D.G. (2013). Fingerprint Based Age Estimation Using 2D Discrete Wavelet Transforms and Principal Component Analysis. International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering, 2(3): 1060-1066. 

[8] Balaji, S., Venkatram, N. (2008). Filtering of noise in fingerprint images. International Journal of Systems and Technologies, 1: 87-94.

[9] Bhavana, D., Ruchi, J., Prakash, T. (2013). Study of fingerprint patterns in relationship with blood group and gender- a statistical review. Res J Forensic Society, 1(1): 15-17. 

[10] Borra, S.R., Reddy, G.J., Reddy, S. (2014). Fingerprint image de-noising using wave atom transform. International Journal of Computer Science Engineering and Information Technology Research, 4: 69-76.

[11] Chen, G.Y., Kegl, B. (2007). Image denoising with complex ridgelets. Pattern Recognition, 40: 578-585.

[12] Devnath, L., Islam, R. (2016). Fingerprint image de-noising by various filters for different noise using wavelet transform. American International Journal of Research in Science, Technology, Engineering & Mathematics, 13(1): 39-44. https://doi.org/10.13140/RG.2.1.2448.3448

[13] Ganesh, B., Dongre, D.S., Jagade, M. (2015). A review and study on fingerprint based gender classification using classified techniques. International Journal of Scientific Research, 6: 596-599. 

[14] Gopinathan, S.K., Thangavel, P.R. (2015). Wavelet and FFT based image de-noising using non-linear filters. International Journal of Electrical and Computer Engineering, 5: 1018-1026. https://doi.org/10.11591/ijece.v5i5.pp1018-1026

[15] Gornale, S.S., Hangarge, M., Pardeshi, R., Kruthi, R. (2015). Haralick feature descriptors for gender classification using fingerprints: A machine learning approach. International Journal of Advanced Research in Computer Science and Software Engineering, 5: 72-78.

[16] Gupta, S., Rao, A.P. (2014). Fingerprint based gender classification using discrete wavelet transform & artificial neural network. International Journal of Computer Science and Mobile Computing, 3(4): 1289-1296. https://doi.org/10.13140/RG.2.2.15292.69769