Advanced Touchless Biometric Identification using SIFT Features

Transcription

1 Advanced Touchless Biometric Identification using SIFT Features H.Abubacker Siddique PG Scholar, Department of CSE National College of Engineering Tirunelveli, Tamilnadu, India. L.K.Indumathi Assistant professor, Dept of CSE National College of Engineering Tirunelveli, Tamilnadu, India Abstract The motto of this paper is to provide an efficient Biometric Identification using Multimodal Touchless Finger Print Acquisition and Mosaicking Technique. Fingerprints are traditionally captured based on contact of the finger on paper or a platen surface. This often results in partial or degraded images due to improper finger placement, skin deformation, slippage and smearing, or sensor noise from wear and tear of surface coatings. A new generation of touchless live scan devices that generate 3D representation of fingerprints is appearing in the market. This new sensing technology addresses many of the problems stated above. Single-view acquisition systems bring in problems, such as scene difference and a limited effective area. This paper thus presents a touchless multiview fingerprint capture system that acquires three different views of fingerprint images at the same time. This device is designed by optimizing parameters regarding the captured fingerprint image quality and device size. A fingerprint mosaicking method is proposed to splice together the captured images of a finger to form a new image with a larger useful print area. In this paper,tscale Invariant Feature Transformation(SIFT) algorithm and Random Sample Consensus techniques are used for better biometric identification. Keywords SIFT, 3D, OTP, Finger Print Recognition, Bio Metrics,Multimodal Bio Metric Systems I. INTRODUCTION Biometrics is an emerging technology that is used to identify people by their physical and/or behavioral characteristics and, so, inherently requires that the person to be identified is physically present at the point of identification. Biometrics is used in computer science as a form of identification and access control. It is also used to identify individuals in groups that are under surveillance. Biometric identifiers are the distinctive, measurable characteristics used to label and describe individuals. Biometric identifiers are often categorized as physiological versus behavioral characteristics. Physiological characteristics are related to the shape of the body. Examples include, but are not limited to fingerprint, face recognition, DNA, Palm print, hand geometry, iris recognition, retina and odour/scent. Behavioral characteristics are related to the pattern of behavior of a person, including but not limited to: typing rhythm, gait, and voice. Some researchers have coined the term behaviometrics to describe the latter class of biometrics. More traditional means of access control include token-based identification systems, such as a driver's license or passport, and knowledge-based identification systems, such as a password or personal identification number. Since biometric identifiers are unique to individuals, they are more reliable in verifying identity than token and knowledgebased methods; however, the collection of biometric identifiers raises privacy concerns about the ultimate use of this information. BIOMETRIC FUNCTIONALITY In Biometric Authentication, many different aspects of physiology,chemistry or behavior can be used. The selection of a particular biometric for use in a specific application involves a weighting of several factors. It is identified seven such factors to be used when assessing the suitability of any trait for use in biometric authentication. Universality means that every person using a system should possess the trait. Uniqueness means the trait should be sufficiently different for individuals in the relevant population such that they can be distinguished from one another. Permanence relates to the manner in which a trait varies over time. More specifically, a trait with 'good' permanence will be reasonably invariant over time with respect to the specific matching algorithm. Measurability (collectability) relates to the ease of acquisition or measurement of the trait. Fig.1:Bio Metric Systems 2014, IJIRAE- All Rights Reserved Page - 158

2 In addition, acquired data should be in a form that permits subsequent processing and extraction of the relevant feature sets. Performance relates to the accuracy, speed, and robustness of technology used (see performance section for more details). Acceptability relates to how well individuals in the relevant population accept the technology such that they are willing to have their biometric trait captured and assessed. Circumvention relates to the ease with which a trait might be imitated using an artifact or substitute. No single biometric will meet all the requirements of every possible application. The block diagram illustrates the two basic modes of a biometric system. First, in verification (or authentication) mode the system performs a one-to-one comparison of a captured biometric with a specific template stored in a biometric database in order to verify the individual is the person they claim to be. Three steps are involved in the verification of a person. In the first step, reference models for all the users are generated and stored in the model database. In the second step, some samples are matched with reference models to generate the genuine and impostor scores and calculate the threshold. Third step is the testing step. This process may use a smart card, username or ID number (e.g. PIN) to indicate which template should be used for comparison. 'Positive recognition' is a common use of the verification mode, "where the aim is to prevent multiple people from using same identity". Second, in identification mode the system performs a one-to-many comparison against a biometric database in attempt to establish the identity of an unknown individual. The system will succeed in identifying the individual if the comparison of the biometric sample to a template in the database falls within a previously set threshold. Identification mode can be used either for 'positive recognition' (so that the user does not have to provide any information about the template to be used) or for 'negative recognition' of the person "where the system establishes whether the person is who she (implicitly or explicitly) denies to be". The latter function can only be achieved through biometrics since other methods of personal recognition such as passwords, PINs or keys are ineffective. The first time an individual uses a biometric system is called enrollment. During the enrollment, biometric information from an individual is captured and stored. In subsequent uses, biometric information is detected and compared with the information stored at the time of enrollment. Note that it is crucial that storage and retrieval of such systems themselves be secure if the biometric system is to be robust. The first block (sensor) is the interface between the real world and the system; it has to acquire all the necessary data. Most of the times it is an image acquisition system, but it can change according to the characteristics desired. The second block performs all the necessary pre-processing: it has to remove artifacts from the sensor, to enhance the input (e.g. removing background noise), to use some kind of normalization, etc. In the third block necessary features are extracted. This step is an important step as the correct features need to be extracted in the optimal way. A vector of numbers or an image with particular properties is used to create a template. A template is a synthesis of the relevant characteristics extracted from the source. Elements of the biometric measurement that are not used in the comparison algorithm are discarded in the template to reduce the filesize and to protect the identity of the enrolee. During the enrollment phase, the template is simply stored somewhere (on a card or within a database or both). During the matching phase, the obtained template is passed to a matcher that compares it with other existing templates, estimating the distance between them using any algorithm (e.g. Hamming distance). The matching program will analyze the template with the input. This will then be output for any specified use or purpose (e.g. entrance in a restricted area)[citation needed]. Selection of biometrics in any practical application depending upon the characteristic measurements and user requirements. We should consider Performance, Acceptability, Circumvention, Robustness, Population coverage, Size, Identity theft deterrence in selecting a particular biometric. Selection of biometric based on user requirement considers Sensor availability, Device availability, Computational time and reliability, Cost, Sensor area and power consumption III.MULTIMODAL BIOMETRIC SYSTEM Multimodal biometric systems use multiple sensors or biometrics to overcome the limitations of unimodal biometric systems. For instance iris recognition systems can be compromised by aging irides and finger scanning systems by worn-out or cut fingerprints. While unimodal biometric systems are limited by the integrity of their identifier, it is unlikely that several unimodal systems will suffer from identical limitations. Multimodal biometric systems can obtain sets of information from the same marker (i.e., multiple images of an iris, or scans of the same finger) or information from different biometrics (requiring fingerprint scans and, using voice recognition, a spoken pass-code). Multimodal biometric systems can integrate these unimodal systems sequentially, simultaneously, a combination thereof, or in series, which refer to sequential, parallel, hierarchical and serial integration modes, respectively. Broadly, the information fusion is divided into three parts, pre-mapping fusion, midstmapping fusion, and post-mapping fusion/late fusion.in pre-mapping fusion information can be combined at sensor level or feature level. Sensor-level fusion can be mainly organized in three classes: (1) single sensor-multiple instances, (2) intra-class multiple sensors, and (3) inter-class multiple sensors. Feature-level fusion can be mainly organized in two categories: (1) intraclass and (2) inter-class. Intra-class is again classified into four subcategories: (a) Same sensor-same features, (b) Same sensordifferent features, (c) Different sensors-same features, and (d) Different sensors-different features. 2014, IJIRAE- All Rights Reserved Page - 159

3 Fig.2: (a) Different views of fingerprint images captured by Surround ImagerTM. (b) Reconstructed 3-D finger shape IV.ALGORITHM Scale-invariant feature transform (or SIFT) is an algorithm in computer vision to detect and describe local features in images. For any object in an image, interesting points on the object can be extracted to provide a "feature description" of the object. This description, extracted from a training image, can then be used to identify the object when attempting to locate the object in a test image containing many other objects. To perform reliable recognition, it is important that the features extracted from the training image be detectable even under changes in image scale, noise and illumination. Such points usually lie on high-contrast regions of the image, such as object edges. Another important characteristic of these features is that the relative positions between them in the original scene shouldn't change from one image to another. For example, if only the four corners of a door were used as features, they would work regardless of the door's position; but if points in the frame were also used, the recognition would fail if the door is opened or closed. Similarly, features located in articulated or flexible objects would typically not work if any change in their internal geometry happens between two images in the set being processed. However, in practice SIFT detects and uses a much larger number of features from the images, which reduces the contribution of the errors caused by these local variations in the average error of all feature matching errors. Lowe's patented method [2] can robustly identify objects even among clutter and under partial occlusion, because his SIFT feature descriptor is invariant to uniform scaling, orientation, and partially invariant to affine distortion and illumination changes. [1] This section summarizes Lowe's object recognition method and mentions a few competing techniques available for object recognition under clutter and partial occlusion. 1) David Lowis method SIFT keypoints of objects are first extracted from a set of reference images [1] and stored in a database. An object is recognized in a new image by individually comparing each feature from the new image to this database and finding candidate matching features based on Euclidean distance of their feature vectors. From the full set of matches, subsets of keypoints that agree on the object and its location, scale, and orientation in the new image are identified to filter out good matches. The determination of consistent clusters is performed rapidly by using an efficient hash table implementation of the generalized Hough transform. Each cluster of 3 or more features that agree on an object and its pose is then subject to further detailed model verification and subsequently outliers are discarded. Finally the probability that a particular set of features indicates the presence of an object is computed, given the accuracy of fit and number of probable false matches. Object matches that pass all these tests can be identified as correct withscale-space extrema detection This is the stage where the interest points, which are called keypoints in the SIFT framework, are detected. For this, the image is convolved with Gaussian filters at different scales, and then the difference of successive Gaussian-blurred images are taken. Keypoints are then taken as maxima/minima of the Difference of Gaussians (DoG) that occur at multiple scales. Specifically, a DoG image is given by, where is the convolution of the original image with the Gaussian blur at scale, i.e., 2014, IJIRAE- All Rights Reserved Page - 160

4 Hence a DoG image between scales and is just the difference of the Gaussian-blurred images at scales and. For scale space extrema detection in the SIFT algorithm, the image is first convolved with Gaussian-blurs at different scales. The convolved images are grouped by octave (an octave corresponds to doubling the value of ), and the value of is selected so that we obtain a fixed number of convolved images per octave. Then the Difference-of-Gaussian images are taken from adjacent Gaussian-blurred images per octave. Once DoG images have been obtained, keypoints are identified as local minima/maxima of the DoG images across scales. This is done by comparing each pixel in the DoG images to its eight neighbors at the same scale and nine corresponding neighboring pixels in each of the neighboring scales. If the pixel value is the maximum or minimum among all compared pixels, it is selected as a candidate keypoint. This keypoint detection step is a variation of one of the blob detection methods developed by Lindeberg by detecting scale-space extrema of the scale normalized Laplacian, [14] that is detecting points that are local extrema with respect to both space and scale, in the discrete case by comparisons with the nearest 26 neighbours in a discretized scale-space volume. The difference of Gaussians operator can be seen as an approximation to the Laplacian, with the implicit normalization in the pyramid also constituting a discrete approximation of the scale-normalized Laplacian. [15] Another real-time implementation of scale-space extrema of the Laplacian operator has been presented by Lindeberg and Bretzner based on a hybrid pyramid representation. [16] 2) Keypoint localization 2014, IJIRAE- All Rights Reserved Page - 161

5 After scale space extrema are detected (their location being shown in the uppermost image) the SIFT algorithm discards low contrast keypoints (remaining points are shown in the middle image) and then filters out those located on edges. Resulting set of keypoints is shown on last image. Scale-space extrema detection produces too many keypoint candidates, some of which are unstable. The next step in the algorithm is to perform a detailed fit to the nearby data for accurate location, scale, and ratio of principal curvatures. This information allows points to be rejected that have low contrast (and are therefore sensitive to noise) or are poorly localized along an edge. Interpolation of nearby data for accurate position First, for each candidate keypoint, interpolation of nearby data is used to accurately determine its position. The initial approach was to just locate each keypoint at the location and scale of the candidate keypoint. [1] The new approach calculates the interpolated location of the extremum, which substantially improves matching and stability. [3] The interpolation is done using the quadratic Taylor expansion of the Difference-of-Gaussian scale-space function, with the candidate keypoint as the origin. This Taylor expansion is given by: where D and its derivatives are evaluated at the candidate keypoint and is the offset from this point. The location of the extremum,, is determined by taking the derivative of this function with respect to and setting it to zero. If the offset is larger than in any dimension, then that's an indication that the extremum lies closer to another candidate keypoint. In this case, the candidate keypoint is changed and the interpolation performed instead about that point. Otherwise the offset is added to its candidate keypoint to get the interpolated estimate for the location of the extremum. A similar subpixel determination of the locations of scale-space extrema is performed in the real-time implementation based on hybrid pyramids developed by Lindeberg and his co-workers. [17] Discarding low-contrast keypoints To discard the keypoints with low contrast, the value of the second-order Taylor expansion is computed at the offset. If this value is less than, the candidate keypoint is discarded. Otherwise it is kept, with final scale-space location, where is the original location of the keypoint. Eliminating edge responses The DoG function will have strong responses along edges, even if the candidate keypoint is not robust to small amounts of noise. Therefore, in order to increase stability, we need to eliminate the keypoints that have poorly determined locations but have high edge responses. For poorly defined peaks in the DoG function, the principal curvature across the edge would be much larger than the principal curvature along it. Finding these principal curvatures amounts to solving for the eigenvalues of the second-order Hessian matrix, H: The eigenvalues of H are proportional to the principal curvatures of D. It turns out that the ratio of the two eigenvalues, say is the larger one, and the smaller one, with ratio, is sufficient for SIFT's purposes. The trace of H, i.e.,, gives us the sum of the two eigenvalues, while its determinant, i.e.,, yields the product. The ratio can be shown to be equal to, which depends only on the ratio of the eigenvalues rather than their individual values. R is minimum when the eigenvalues are equal to each other. Therefore the higher the absolute difference between the two eigenvalues, which is equivalent to a higher absolute difference between the two principal curvatures of D, the higher the value of R. It follows that, for some threshold eigenvalue ratio, if R for a candidate keypoint is larger than, that keypoint is poorly localized and hence rejected. The new approach uses. [3] This processing step for suppressing responses at edges is a transfer of a corresponding approach in the Harris operator for corner detection. The difference is that the measure for thresholding is computed from the Hessian matrix instead of a second-moment matrix (see structure tensor). 2014, IJIRAE- All Rights Reserved Page - 162

6 3) Orientation assignment In this step, each keypoint is assigned one or more orientations based on local image gradient directions. This is the key step in achieving invariance to rotation as the keypoint descriptor can be represented relative to this orientation and therefore achieve invariance to image rotation. First, the Gaussian-smoothed image at the keypoint's scale is taken so that all computations are performed in a scaleinvariant manner. For an image sample at scale, the gradient magnitude,, and orientation,, are precomputed using pixel differences: The magnitude and direction calculations for the gradient are done for every pixel in a neighboring region around the keypoint in the Gaussian-blurred image L. An orientation histogram with 36 bins is formed, with each bin covering 10 degrees. Each sample in the neighboring window added to a histogram bin is weighted by its gradient magnitude and by a Gaussian-weighted circular window with a that is 1.5 times that of the scale of the keypoint. The peaks in this histogram correspond to dominant orientations. Once the histogram is filled, the orientations corresponding to the highest peak and local peaks that are within 80% of the highest peaks are assigned to the keypoint. In the case of multiple orientations being assigned, an additional keypoint is created having the same location and scale as the original keypoint for each additional orientation. 4) Keypoint descriptor Previous steps found keypoint locations at particular scales and assigned orientations to them. This ensured invariance to image location, scale and rotation. Now we want to compute a descriptor vector for each keypoint such that the descriptor is highly distinctive and partially invariant to the remaining variations such as illumination, 3D viewpoint, etc. This step is performed on the image closest in scale to the keypoint's scale. First a set of orientation histograms is created on 4x4 pixel neighborhoods with 8 bins each. These histograms are computed from magnitude and orientation values of samples in a 16 x 16 region around the keypoint such that each histogram contains samples from a 4 x 4 subregion of the original neighborhood region. The magnitudes are further weighted by a Gaussian function with equal to one half the width of the descriptor window. The descriptor then becomes a vector of all the values of these histograms. Since there are 4 x 4 = 16 histograms each with 8 bins the vector has 128 elements. This vector is then normalized to unit length in order to enhance invariance to affine changes in illumination. To reduce the effects of non-linear illumination a threshold of 0.2 is applied and the vector is again normalized. Although the dimension of the descriptor, i.e. 128, seems high, descriptors with lower dimension than this don't perform as well across the range of matching tasks [3] and the computational cost remains low due to the approximate BBF (see below) method used for finding the nearest-neighbor. Longer descriptors continue to do better but not by much and there is an additional danger of increased sensitivity to distortion and occlusion. It is also shown that feature matching accuracy is above 50% for viewpoint changes of up to 50 degrees. Therefore SIFT descriptors are invariant to minor affine changes. To test the distinctiveness of the SIFT descriptors, matching accuracy is also measured against varying number of keypoints in the testing database, and it is shown that matching accuracy decreases only very slightly for very large database sizes, thus indicating that SIFT features are highly distinctive. V.SYSTEM IMPLEMENTATION This paper consists of the modules such as Image Acquisition, Image Preprocessing, Ridge Area selection, Minutiae and SIFT Feature, Over lapping for Mosaick Image, Stitching Point, Authentication IMAGE ACQUISITION It is designed to place the finger with fixed position. Schematic diagram of the proposed touchless multi view fingerprint acquisition device. 2014, IJIRAE- All Rights Reserved Page - 163

7 IMAGE PREPROCESSING Histogram :wanted/unwanted area separation using 0 s and 1 s from the matric form by image intensity value. color gray, gray color ( range 0 to 255) Binarization-To separate an object from the background Fig: Image Preprocessing RIDGE AREA SELECTION From the converted matrix form, the value1 s areas are considered for security. High weighted area is considered otherwise reject the Area Fig: Binary Image Fig: Ridge Area MINUTIAE POINT The selected ridge area basically in matrix form.there the first row and column is deleted to find minutiae's by thin plate spline algorithm condition. Minutiae : ridge point start and end inside the ridge area. 2014, IJIRAE- All Rights Reserved Page - 164

8 Fig: Minutia Point SIFT FEATURE Fig.:Obtaining SIFT features OVERLAPPING FOR MOSAICK IMAGE Fig:Overlapping for Mosaick Image STITCHING stitching points are detected for generate the pattern value for authentication From the overlapped image the active points are stitched to generate the pattern. The stitching line by comparing the similarity of overlapping region. Stitching the multiple fingerprint images captured from different views into one single fingerprint image Fig.: Stitching output AUTHENTICATION The generated pattern value which is matched with database image pattern value Fig.: Authentication The pattern value based on the stitching points intensity value matrix with the supporting of threshold to authenticate the persons with the database set. 2014, IJIRAE- All Rights Reserved Page - 165

9 VI.PERFORMANCE ANALYSIS MATCHING FACTOR VII.REFERENCES [1] S. Pankanti, S. Prabhakar, and A. Jain, On the individuality of fingerprints, IEEE Trans. Pattern Anal. Mach. Intell., vol. 24, no. 8, pp , Aug [2] X. Xia and L. O Gorman, Innovations in fingerprint capture devices, Pattern Recognit., vol. 36, no. 2, pp , Feb [3] Y. Song, C. Lee, and J. Kim, A new scheme for touchless fingerprint recognition system, in Proc. Int. Symp. Intell. Signal Proces. Commun.Syst., Nov. 2004, pp [4] Mitsubishi. Mitsubishi Touchless Fingerprint Sensor [Online]. Available: [5] Lumidigm. Lumidigm Multispectral Fingerprint Imaging [Online].Available: [6] Distinctive image features from scale-invariant keypoints. David G. Lowe, International Journal of Computer Vision, 60, 2 (2004). 2014, IJIRAE- All Rights Reserved Page - 166