A Novel Technique for Principal Lines Extraction in ...

3 downloads 0 Views 651KB Size Report
Corresponding Author: Zia-uddin, Department of Computer Science, Islamia College Peshawar ... Zia-uddin, Zahoor Jan, Jamil Ahmad and Almas Abbasi. 1. 1. 1.
World Applied Sciences Journal 31 (12): 2010-2014, 2014 ISSN 1818-4952 © IDOSI Publications, 2014 DOI: 10.5829/idosi.wasj.2014.31.12.229

A Novel Technique for Principal Lines Extraction in Palmprint Using Morphological TOP-HAT Filtering 1

Zia-uddin, 1Zahoor Jan, 1Jamil Ahmad and 2Almas Abbasi

Department of Computer Science, Islamia College Peshawar (Public Sector University), University Campus, Peshawar, Khyber Pakhtunkhwa, Pakistan 2 Faculty of Computer Science and Information Technology, University of Malaya, Malaysia 1

Abstract: Palmprint is one of the main biometric features that has gained a lot of attention recently due to its commercial applications. Palmprint is more flexible in identifying an individual based on wrinkles, ridges and Principal lines. These features do not change during the lifetime of a person and no two individuals have similar palmprints. Most of the conventional methods use edge detection operators to detect lines in the palm which produce too many trivial lines. In this paper, a novel technique has been proposed to extract principal lines in two phases without using edge detection. The first phase involves some preprocessing operations and the second phase consists of morphological operations and elimination of undesirable components. The experimental results show that the proposed technique is more effective than existing techniques. Key words: Features Extraction

Palmprint recognition

INTRODUCTION The palm area of hand contains Principal lines, wrinkles and ridges. The major lines are developed between the 3rd and 5th month of pregnancy [1] and the remaining lines develop subsequently. Palmprints are not heredity [2] and even twins don't possess identical Palmprints. Palmprint recognition is receiving more attention of the researchers because of simplicity in image acquisition using inexpensive imaging devices, uniqueness, permanence, user approval and cost efficiency of the system. Palmprint recognition works with both high resolution and low resolution images. The former images are used for legal applications by law enforcing organizations and the later are suitable for public and commercial applications. High resolution images are more suitable for the extraction of minutia points, singular points and ridges, while low resolution images are ideal for principal lines, wrinkles and textures extraction [3]. Initial research work focused on legal based applications Corresponding Author:

Biometrics

Morphology

Top-Hat filter

using high resolution images but recently the trend is changing and most of the research is dedicated to commercial applications using low resolution images. The success of a Biometrics system depends on five factors: accuracy, computation speed, security, cost and user acceptance and environment constraints [3]. Palmprint recognition algorithms include line, texture/subspace and statistics based. The line approach depends on information of Principal lines like edge point detection, length and orientation [4]. Texture algorithms use Principal lines, ridges and wrinkles and invoke transform like Gabor, Discrete Cosine Transform (DCT), Fast Fourier Transform (FFT), Discrete Wavelet Transform (DWT) etc. independently [5, 6] or combined using statistical operators. While the statistical based algorithms utilize statistical quantities like mean, standard deviation, median [7, 8], etc for the feature vector extraction. This paper presents a novel technique for the extraction of principal lines that can be used in palmprint recognition. The following section of the paper consists

Zia-uddin, Department of Computer Science, Islamia College Peshawar (Public Sector University) University Campus, Peshawar, Khyber Pakhtunkhwa, Pakistan.

2010

World Appl. Sci. J., 31 (12): 2010-2014, 2014

of literature review of some of the state of the art techniques, followed by the proposed technique. Results and conclusions are given in the end. Related Work: Tunkpien et al, first images were blurred using a smoothing filter before any edge detection. Horizontal and vertical lines were detected and combined from a 2x2 gradient operator which was calculated in two directions [4]. Han et al. extracted palm lines from region of interest by using Sobel edge operators [7]. Lines were enhanced using different morphological procedures. The resultant image was split into three different shape sizes. Feature vector was obtained by calculating mean of pixels in all blocks. Sakdanupab et al. first normalized image to reduce the effects of different brightness [9]. The gray level profiles in a small window of size 3x3 at 0o, 45o, 90o and 135o degree were used for principal line detection. Noise is removed using morphological closing and smoothing is applied to the resultant image. Zhu et al. extracted coinciding Principal lines by calculating gradient image of four directions [10]. Filters were applied to the resultant image. Using AND operator, filtered image and edges extracted using canny edge function were combined.. Principal lines were extracted by cascading different morphological methods. The feature vector is calculated using shape matching with the help of nearest neighbour. The final recognition was accomplished using Cosine similarity. Wu et al. extracted Principal lines using canny edge operator. The scale and direction of edges were determined [11]. Four groups were formed using edges, each indicating edges in a specific direction. Feature vector was calculated from derivation of fuzzy energy in each direction. Leung et al. lines were extracted using sobel edge function [12]. Trivial lines were removed using thresholding and feature vector was extracted depend on the remaining lines. MATERIALS AND METHODS Most of the researchers proposed edge detection operator such as Sobel, Canny, Prewitt etc. for Principal lines detection, which produce too many trivial lines. The proposed line extraction method does not use conventional edge detection approach instead it uses two phase approach as illustrated in Fig. 1. Preprocessing Phase: In the preprocessing phase, the contrast of the palm image is enhanced by normalization. Noise is removed with the help of smoothing and median

Fig. 1: Block diagram of proposed technique

Fig. 2: Palm Image (a) Orignal Image (b) normalized image using L. Hong et al technique [11]. filters. Palm image is converted to negative in the last step of the phase. The detailed procedure of the Preprocessing phase is as follows: Step 1: Palm images with uneven contrast are enhanced with the enhancement technique in equation (a) and (b) proposed by L. Hong et al. [11]. Fig. 2 show the original and contrast enhanced images.  I ′ ( i, j ) =   =

d

+

ifI ( i. j ) >

d



otherwise

d ( I (i,

j ) − )2

(a)

(b)

where d is standard deviation and d is the average of the output image and I'(i, j) is the normalized image. is is the average of the input standard deviation and image. 2011

World Appl. Sci. J., 31 (12): 2010-2014, 2014 Table 1: 5x5 smoothing mask filter 1/25 1/25 1/25 1/25 1/25

Fig. 3: Palm image after applying (a) smoothing filter and (b) standard median filter

1/25 1/25 1/25 1/25 1/25

1/25 1/25 1/25 1/25 1/25

1/25 1/25 1/25 1/25 1/25

1/25 1/25 1/25 1/25 1/25

Step 3: After normalization and removal of noise, the resultant enhanced image is converted to negative in order to have the region of interest in higher intensities as shown in Fig. 4. Extraction Phase: In Extraction phase Top-Hat filter and contrast enhancement applied respectively on the negative image. Noise is removed with connected component labeling after binarization. The detailed procedure of the extraction phase is as follow: Step 1: We propose to correct uneven illumination by using Top-Hat filter. Top-hat filtering calculates the morphological opening and then subtracts the resultant image from the source image shown in eq. (c).

Fig. 4: Palm image is converted to Negative image

(c)

T=AWhere = (A

Fig. 5: Top-Hat Filter is applied to negative image

)

r

morphological Where A denotes original image, erosion and dilation opening, structure element, respectively Keeping in view, Top-Hat filter with a disk-shaped structuring element with 2 point radius is used to remove the uneven background illumination from an image. The resultant image is show in Fig. 5. Step 2: Linear Contrast adjustment is applied by scaling the intensity values in the original image to improve the visibility of the image. Fig.6 shows the resultant image after contrast adjustment.

Fig. 6: Contrast of the resultant image from Top-Hat filter is adjusted

Step 3: The grayscale image resulted from Top-Hat filter and contrast adjustment is converted to a binary image using a threshold value. The result is shown in Fig. 7.

Step 2: Noise and trivial lines are removed by smoothing the image. A smoothing filter of 5x5 standard mask shown in Table 1 is applied to blur the image. Standard median filter is applied to remove any leftover noise. Results show in Figure 3.

Step 4: Connected component labeling is used for trivial lines removal. 8-neighborhood is used for identifying connected components. Components which have less pixels than threshold value are usually trivial lines and hence are removed. The final image after removal of noise is shown in Fig. 8. 2012

World Appl. Sci. J., 31 (12): 2010-2014, 2014

Fig. 7: Palm Image

image

is

converted

to

Binary

RESULTS AND DISCUSSION IIT Palmprint Database version 1.0 [13] is a widely used database for palmprint research. The current database consists of images of 235 users. 100 random images are selected for this research. The proposed technique was compared with Tunkpien et al [4], Sakdanupab et al. [9] and Zhu et al. [10] in terms of visual results, execution time and accuracy. Figure 9 shows that show that the proposed technique produce less trivial lines and as a result create more vibrant principal lines as compared to those methods (a)

(b)

Fig. 8: Final Palm image after noise removal using connected component labeling which use edge detection operator. The results in tables 2 also show that the proposed technique have high accuracy rate compared to all three techniques where as less execution time than Sakdanupab et al. [7] and Zhu et al. [8]. Table 2: Comparision of execution time and accuracy of diferent line extraction techniques Method Execution time Accuracy Proposed method 0.0230 ms 93.33% Tunkpien et al [4 ] 0.0197 ms 46.66% Sakdanupab et al. [7] 0.0242 ms 50% Zhu et al. [8] 0.0509 ms 65%

(c)

(d)

(e)

Fig. 9: Experimental results of (a) Original image (b) Proposed Image (c) Zhu et al. [8] (d) Tunkpien et al [4 ] (e) Sakdanupab et al. [7] 2013

World Appl. Sci. J., 31 (12): 2010-2014, 2014

CONCLUSION

6.

This paper proposes a novel technique. The proposed technique is simple and effectively extracts Principal lines from Palmprint without using conventional edge detection operator. Results obtained from the experiments clearly show that the proposed technique is more accurate and produces far better results than edge detection based techniques.

7.

8.

REFERENCES 1.

2.

3.

4.

5.

Wu, X., K. Wang and D. Zhang, 2002. "Line feature extraction and matching in palmprint," in Second International Conference on Image and Graphics, pp: 583-590. Kong, A.W.K., D. Zhang and G. Lu, 2006. "A study of identical twins' palmprints for personal verification," Pattern Recognition, 39: 2149-2156. Kong, A., D. Zhang and M. Kamel, 2009. "A survey of palmprint recognition," Pattern Recognition, 42: 1408-1418, 7. Tunkpien, P., S. Panduwadeethorn and S. Phimoltares, 2010. "Compact extraction of principle lines in palmprint using consecutive filtering operations," Advanced Virtual and Intelligent Computing (AVIC) research centerDepartment of Mathematics, Faculty of Science, Chulalongkorn University,(July, 24-25, 2010), 2010. Imtiaz, H. and S.A. Fattah, 2010. "A DCT-based feature extraction algorithm for palm-print recognition," in Communication Control and Computing Technologies (ICCCCT), 2010 IEEE International Conference on, pp: 657-660.

9.

10.

11.

12.

13.

2014

Jan, Z. and A.M. Mirza, 2012. "Genetic programming-based perceptual shaping of a digital watermark in the wavelet domain using Morton scanning," Journal of the Chinese Institute of Engineers, 35: 85-99. Han, C.C., H.L. Cheng, C.L. Lin and K.C. Fan, 2003. "Personal authentication using palm-print features," Pattern Recognition, 36: 371-381. Ullah, H., Z. Jan, R.J. Qureshi and B. Shams, 2013. "Automated Localization of Optic Disc in Colour Fundus Images," World Applied Sciences Journal, 28: 6. Sakdanupab, M. and N. Covavisaruch, 2008. "An efficient approach for automatic palmprint classification," in Signal Image Technology and Internet Based Systems, 2008. SITIS'08. IEEE International Conference on, pp: 229-234. Zhu, L. and R. Xing, 2009. "Hierarchical palmprint recognition based on major line feature and dual tree complex wavelet texture feature," in Fuzzy Systems and Knowledge Discovery, 2009. FSKD'09. Sixth International Conference on, pp: 15-19. Wu, X., K. Wang and D. Zhang, 2004. "An approach to line feature representation and matching for palmprint recognition," Journal of Software, 15: 869-880. Leung, M.K., A.C.M. Fong and S.C. Hui, 2007. "Palmprint verification for controlling access to shared computing resources," Pervasive Computing, IEEE, 6: 40-47. Kumar, A., 2008. "Incorporating cohort information for reliable palmprint authentication," in Computer Vision, Graphics & Image Processing, 2008. ICVGIP'08. Sixth Indian Conference on, pp: 583-590.