Chethan It provides fast and efficient data structures

Chethan Kumar Hanumanthappa Venkatesh (Author)Information Technology,SRH Hochschule HeidelbergHeidelberg, GermanyMail ID: [email protected] Abstract— This paper is to show Touch-less fingerprint recognition is regarded as aviable alternative to contact-based fingerprint recognition technology. Itprovides a near ideal solution to the problems in terms of hygienic,maintenance and latent fingerprints. In this paper, we present a touch-lessfingerprint recognition system by using a digital camera. This paper is to showthe skin tone segmentation using open CV and C++programming language and colordetection with fingerprint detection of hand in different tilted positions offingers. And fingerprint detection with inclusion of hand palm.

Keywords: Luminance, colourtransform, skin tone detection.                                                                                                                                                       I.         IntroductionThe touch-based electronic fingerprint scanner will lead to theweakening of durability if the device is used heavily.

Best services for writing your paper according to Trustpilot

Premium Partner
From $18.00 per page
4,8 / 5
Writers Experience
Recommended Service
From $13.90 per page
4,6 / 5
Writers Experience
From $20.00 per page
4,5 / 5
Writers Experience
* All Partners were chosen among 50+ writing services by our Customer Satisfaction Team

In addition, thepressure of the physical contacts will normally cause the touch-basedfingerprint images to be degraded. On the other hand, images captured withtouch-less devices are distortion free and present no deformation because theseimages are free from the pressure of contact. Moreover, the problems in termsof hygienic, maintenance, latent fingerprint problems and so forth are overcomewith the touch-less fingerprinting technology 2. Finger print detection willbe helpful in various day to day applications like human machine interface,surveillance, driver detection, Immigration, security and access in criticalareas. In this project we should use a built-in laptopcamera or a web-camera. For a value to luminance to brighten the image we canuse either of the two alternations of the RGB (Red, Green, Blue) color model.

Either HSI (Hue, Saturation, Lightness) or HSV (Hue, Saturation, Value). Thebelow flow diagram will help us to understand the concept of touchlessfingerprint capturing technology.         II.TECHNOLOGIES INVOLVED A.    OpenCV                We are using the OpenCV (OpenSource Computer Vision Library) which is a library mainly aims at real-timecomputer vision systems, developed by Intel.

It mainly focuses on real timeImage Processing and Computer Vision. Computer Vision is a science andtechnology that is used to process images to obtain relevant information.OpenCV library is written in C language with its primary application interfaceavailable in C.

It provides wrapper classes for use in C++ language as well. Itprovides basic data structures for Matrix operations and Image Processing withefficient optimizations. It provides fast and efficient data structures andalgorithms which can be used for complex real time Image Processing andComputer Vision applications 3. In general, the conversion of image from 2Dto 3D is done in MATLAB simulation, however, we use QT creator to process realtime image that programs finger print detection.  B.    C++Theprogramming language used is C++ (pronounced as see plus plus), it is amulti-paradigm, general-purpose and compiled programming language. Whichcomprises of a combination of high-level language features like OOPS concepts andlow-level language features like memory access using pointers etc. It has also addedobject oriented concepts like classes to the C language.

We are using here theC++ interface of OpenCV for the implementation. The C++ interface provides apowerful data structure Mat (short for Matrix) which is efficient, portable andeasy to use. Mat is basically just a multi-channel and multi-dimensional arraywhich can be used to store images, intermediate image transformations, valuesjust like a normal Array. III.IMPLEMENTATIONA.

    Skintone SegmentationThe Skin tonesegmentation process deals with separating the user’s skin from the backgroundin the image. This can be done using so many methods. The purpose of skin tone detectionis to determine skin pixels in the image and generate skin region bydiscriminating skin and non-skin pixels. There will be noise in the segmented skintone. Ground subtraction method used to eliminate background noise. The thresholdingstep is used to separate the skin from the background. Thresholding used toextract an object from its background by calculating its intensity values for eachpixel suchthat eachpixel is either classified as anobject pixel or background pixel.

Thresholding is done on the input imageaccording to a threshold value. The object in use is the user’s hand. The mostimportant component for thresholding is the threshold value. The other type of applyingthresholding values for better result explained in the further points 5. HandSegmentation method with background constraints explained as follows: 1)       With Background Constraint  Using this technique the noise can redused largely by puttingbonstrains at background to extarct the hand blob. The intensity of pixels are usedfor user hand segmentation. Usually the hand intencity is much higher, so bykeeping background dark, hand can be easily segmented. This type includes thefollowing methods for segmentation:  a)       Static Threshold Value  Inthis type of thresholding the image frame is taken as inputfrom the webcam/inbuild laptop camera in the RGB format and it is convertedinto grayscale.

After that either the static threshold value is used or athreshold value is selected from 0 to 255, which acts as the threshold value. Thisthreshold value should be considered such that the white blob of hand is  segmented with minimum nosi possible. We canuse a trackebar to adjust the threshold value for the image. In this methodthe user have to give a proper threshold value to get the exact segmentationrequired.

Hence this method it get complicated than the. So the success or thefailure of this segmentation is in the hand of user, the quality ofsegmentation may be low.Whenever there is intencity variation in the hand thesegmentation will be poor. It is very importatnt to have a constant background inthis method.However the constant lighting conditions during every system use,the system might fail depending on the user’s hand color. If the user hand isunable to reflect the light light properly(in case of dark colour.), the systemmight not be able to separate the user’s hands and the dark background. Thefigures 2 below show the hand segmentation without the noise.

The noise found infigure 1 clearly shows how the user is unable to segment the hand (a badthresholding value) which can reduce the accuracy of hand detection.The region ofinterest (ROI) in our case is user hand. Which is needed to extract fromoriginal images which makes the work faster and reduces the computational timetaken. The HSV color space based skin filter was used to form the binarysilhouette of the input image, which will be used to segment hands and maskingwith original images. Essentially, HSV type color spaces are deformations ofthe RGB color cube and they can be mapped from the RGB space via a nonlineartransformation. The reason behind the selection of this color space in skindetection is that it allows users to intuitively specify the boundary of theskin color class in terms of the hue and saturation.

As value in HSV providesthe brightness information, it is often dropped to reduce illuminationdependency of skin color.  Fig.1 Hand segmentation with noise.  Fig.2. Hand segmentation without noise.

 b)      Incremental Thresholding Value  In this method,same pre-processing as in the static thresholding value is done on the imageinput frame, converting from RGB to Grayscale. Instead of using a constantvalue for every input image frame, the threshold value is incremented till a conditionis not met. For this method, a minimum threshold value is set and then theinput image frame is thresholded using this value. If the current thresholdingvalue does not fulfil the condition, then the thresholding value is incrementedand again the same procedure is followed till the condition is met.

Thecondition to detect hand is until only one white blob is present in thethresholded image.  c)       Thresholding using Otsu’s Method  Otsu’s Methodisused to automatically select a threshold value based on the shape of thehistogram of the image. This algorithm assumes that the image contains twodominant peaks of pixel intensities in the histogram that is two classes ofpixels. The histogram should be bi-model for using this method. The two classesare foreground and background. The algorithm tries to find the optimalthreshold value using which the two classes are separated in such a way thattheir intra-class variance or combined spread is minimal.

Thus the two classesare background and hand. So the threshold value tries to separate the peaks inorder to give minimal intra-class variance4. d)       Dynamic Thresholding using Color at Real Time  Unlike previousmethods of thresholding, in this method color based thresholding is done. Thiscan also be termed as color level slicing. Initially the user has to give somedummy input image frames with the hand to be detected in the central part ofthe image. The system would do the analysis on these dummy input frames andgenerating dynamic threshold values in RGB. In this analysis, a small centralcircular part, with arbitrary radius, of the dummy input frames is consideredinitially. The first two pixels of the central part are set as minimum pixelvalue and maximum pixel values.

Then rest all pixels in the central part areprocessed. For every pixel value that is scanned, it is compared with theminimum and maximum pixel values. If the scanned pixel value is less than theminimum pixels value then the minimum pixel value is updated to the scannedpixel value. Similarly if scanned pixel value is more than that of maximumpixel value, then the maximum pixel value is updated to the scanned pixelvalue. The range defined by the Minimum and Maximum pixel value is used tothreshold the image, whichever pixel comes between this ranges is considered ashand pixel.B.

    Human Skin ModellingComputer visionrequires colour transformations, there are several ways including, RGB, HSV,XYZ, LHC etc., The RGB represents the view of colours in proportions of Red,Green & Blue. 6 1)      RGBThis isa simple yet powerful method to construct a skin classifier directly from theRGB composites which sets many rules (N) for skin colour likelihood. e.

g.,luminance elimination. They utilize the following rules: An R, G, B pixelis classified as skin if and only if  R > 95 & G > 40 & B > 20 & max (R, G, B)- min (R, G, B) > 15 (4) & |R-G|> 15 & R > G & R > B Some authorsprefer to normalise the RGB primaries beforehand. Let the RGB denote thenormalisedcolour space, which is expressed below The bcomponent has the least representation of skin colour and therefore it can benot taken into count in skin segmentation 7. Abdullah-Al-Wadud and Chae 8use a colour distance map (CDM) applied to RGB colours, although that can beextended to any colour space. They have implement an algorithm based on theproperty of the flow of water to further refine the output using an edgeoperator.

The generated CDM is a grayscale image. The distribution of thedistance map is quasi-Gaussian. They also propose an adaptive Standard SkinColour (SSC) to act as a classifier to vote for skin pixels. The method doesnot develop any color space. 2)       HSV HSV(Hue, Saturation and Value) colour space. HS can be obtained by applying a non-lineartransformation to the RGB colour primaries as shown in below equation. Atexture amplitude map is used to find regions of low texture information.

Thealgorithm first locates images containing large areas where colour and textureis appropriate for skin, and then segregates those regions with little texture.The texture amplitude map is generated from the matrix I by applying 2D medianfilters. RGB to HSV transform can be expressed as in-                                                                                                            1                                                                                            C.     Finger print extractionTo extract fingerprint from the palm we are going to use the convex hall technique. Convex Hullis the smallest convex set that contains a set of points.

And a convex set is aset of points such that, if we trace a straight line from any pair of points inthe set, that line must be also be inside the region. With this algorithm wecan get much smoother, nice & easier analyse. 9.  The fingertipangle can be calculated by following equation.||X|| = {(X – X1)(X – X2)} / {|X –X1| |X-X2|} X1, X2 are the previous and following point of X and they are separated byX within the same distance. The value of fingertip-angle ||X|| is between 0 and1, because it is the cosine value of ‘X1 X X2’. We notice that more accurate theangle is, the larger the cosine value is. Thus all the angle of each point incontour is calculated and a threshold is set for cosine value which givesfingertip position.

 By writing arectangle to the fingertip we can get the finger print within the rectangle. Usingyou can find convexity defects, which will be potential fingertip locations.Following figures shows the fingertip extraction for multiple fingers.  D.   Finger Detection and tiltingWith palm andfinger segmentation been done, we can implement the same method to visualizeonly fingers segmentation.

We can make the segmented image of finger to rotateat the desired angle of 30 degrees. In this we sample a the pixels and thenrotate or tilt the finger. This can be achieved by considering the points fromthe finger as sample point (A, B) as coordinates. Respective equations for ‘x’and ‘y’ are determined in terms of cosine and sine with an angle 180 degrees.

Equationfor x and y axis are,x = cos (angle * ?÷ 180) + Xy = sin (angle * ?÷ 180) +YWhen detectingsingle or multiple fingers, this does have noise, we can eliminate it bychanging HSV bound values accordingly 10.RESULTSThe current workcan efficiently result in extraction of segmented skin tone and thefingerprint. Skin tone segmentation is provided in different techniques for preprocessing ofinput images. And the fingerprint obtained can be used in several applicationsas mentioned in introduction. These provided techniques can be used accuratelyand efficiently applying some constraints.

This method can also be used forhuman machine interfaces. ConclusionThe paper is presented to get the handsegmentation & fingertip print extraction to the user with the hand palm.By changing the values, you must adjust the segmentation and the fingertipdetection. This algorithm helps in getting the fingerprint without any physicalcontact of fingers. This technique can be used in multiple areas. By using thisalgorithm, you can get more durable, efficient and fast fingerprint detection.To detect the hand palm & fingers a very well illuminated area isnecessary.  With a good backgroundsubtraction, we can use this method in all locations.

 AcknowledgmentThis paper iswritten under the guidence of lectures from Prof. Dr. Ing Christof Jonietz and from the referencesgiven in the paper. I thank all the authors and Prof. Dr. Ing Christof Jonietz for giving meopprtunity to write this paper.REFERENCES1     A Skin Tone Detection Algorithmfor an Adaptive Approach to  Steganography,by Abbas Cheddad, Joan Condell, Kevin Curran and Paul Mc Kevitt2     Y.

Song, C. Lee, and J. Kim,”A New Scheme for Touchless Fingerprint Recognition System”,International Symposium on Intelligent Signal Processing and CommunicationSystems, Korea, 2004.3     O’reilly, Learning OpenCV,Computer Vision in C++ with the OpenCV Library, Adrian Kaebler and GaryBradski.4     Implementation of HandDetection based Techniques for Human Computer Interaction, by Amiraj Dhawan,Vipul Honrao5     Thresholding

pdf6      G. Gomez,On selecting colour components for skin detection, in: Proceedings ofInternational Conference on Pattern Recognition, Quebec, 11-15 Aug 2002, vol.2,pp. 961-964.

7      R.R.Porle, A. Chekima, F. Wong and G. Sainarayanan, Wavelet-based skin segmentationfor detecting occluded arms in human body pose modelling system, in: Proceedingsof International Conference on Intelligent and Advanced Systems, Malaysia,25-28 November 2007, pp.

764 -769.8       M. Abdullah-Al-Wadud and O. Chae, Skinsegmentation using color distance map and water-flow property, in: Proceedingsof International Conference on Information Assurance and Security, Italy, 8-10Sept. 2008, pp.



I'm Simon!

Would you like to get a custom essay? How about receiving a customized one?

Check it out