Segmentation and Detection of Iris Region for Noisy Regions Taken in Unconstrained Using Fuzzy Weightage Membership Function and Hough Transform

For more strict security requirements, biometric research has experienced significant advances in recent years. More important is the need to overcome the rigid constraints necessitated by the practical implementation of sensible but effective security methods such as recognition. An inventive iris acquisition method with less constraints on the iris verification and identification process as well as on the subject. Consequently, to provide acceptable measures of accuracy, it is critical for such an iris recognition system to be complemented by a robust iris segmentation approach to overcome various noise effects introduced through image capture under different recording environments and scenrarios. The proposed algorithm consists of few steps of detecting the approximate localization of the eye area of the noisy image captured at the visible wavelength using the extracted sclera area, defining the outer iris boundary which is the boundary between iris and sclera detecting the upper and lower eyelids, conducting the verification and correction for outer iris boundary detection and detecting the pupil area and eyelashes and conducting the verification of the reliability of the segmentation results. Experimental results clear that it gives better segmentation of iris region and better detecting of upper and lower eyelids which result in valid iris region detection. This can be implemented in real time application due to less time taken by the algorithm in segmentation. Keywords— Iris detection, Segmentation, Hough transforms I. BIOMETRIC TECHNOLOGY A biometric system provides automatic recognition of a person on the basis of some sort of unique feature or characteristic possessed by that individual. Biometric systems have been developed on the basis of fingerprints, facial features, voice, hand geometry, handwriting, the retina and the one that is presented in this paper i.e. the iris. Biometric systems work by capturing a sample of the feature first, for example recording a digital sound signal for voice recognition. After taking the sample, that sample is transformed using some kind of mathematical function into a biometric template. The biometric template will then give a normalized efficient and highly discriminating representation feature which can then be considerately compared with other templates in order to determine identity. Most biometric systems allow two modes of operation. An enrolment mode for the addition of templates to a database of preen rolled templates. A good biometric is described by use of a feature that is highly distinctive so that the chance of any two people having the same characteristic will be minimal, stable-so that the feature does not change over time, can be easily capturedin order to provide ease to the user, and prevent falsehood. II. IRIS AS A BIOMETRIC TOOL The iris is a thin circular diaphragm which lies between the lens and the cornea of the human eye. The iris is breached nearer to its center by a circular aperture known as the pupil. The use of the iris is to control the amount of light entering through the pupil and it is done by the sphincter and the dilator muscles which alter the size of the pupil. The average radius of the iris is 6mm, and the pupil size can vary from 10% to 80% of the iris radius. The iris contains a number of layers, the lower is the epithelium layer, which contains thick pigmentation cells. The stromal layer is above the epithelium layer and consists of blood vessels pigment cell and two iris muscles. The density of stromal pigmentation tells the colour of the iris. The externally visible surface of the multi layered iris consists of two zones, which slightly differ in colour. An external ciliary zone and these two zones are divided by the collaretswhich is shown as a zigzag pattern. The development of the iris starts during the third month of embryonic life. The distinctive pattern on the iris is made during the first few years. Development of the distinctive patterns of the iris is random and not related to any genetic factors. The only feature that depends on genetics is the pigmentation of the iris, which determines its colour. Due to the epigenetic nature of iris patterns, the two eyes of a person consist of completely independent iris patterns and identical twins possess uncorrelated iris patterns. International Journal of Emerging Technology and Advanced Engineering Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 4, Issue 7, July 2014) 960 III. PRESENT SCHEME A state of art iris recognition system comprises the following four basic modules: Image acquisition  Iris segmentation  Normalization and iris code generation  Comparison and recognition A typical iris segmentation process includes the following steps: demarcating the iris inner and outer boundaries at the pupil and sclera; demarcating its upper and lower eyelids if they occlude; and detecting and excluding any superimposed occlusions such as a eyelash, eyelids, shadows, or reflections. We have presented a method which takes less time for segmentation and finds accurate circles for pupil and iris. This technique uses median filters to enhance the image before segmentation. Thus segmentation is applied which divides the image into a number of clusters only, in which iris region belongs to single cluster only. After that edge detection has been carried out in order to find better edges needed in hough transform for finding iris circle. Below is the proposed algorithm in this work. Step IPreprocess the input image in order to extract eye region from the input image. Filters and extract eye region from the input image. Filters and extract eye region from the input image. Filter and contrast adjustment functions are used in this step to better highlight eye region. Step IISmoothing filters have been applied on the extracted eye portion in order to get better results in image segmentation. Step IIISegmentation has been done for the extracted portion using fuzzy means clustering and iris region has been located in a cluster. Step IVEdge detection has been applied on the segmented image. We tried different filters like Sobel, Prewitt, Canny but Canny gave better results in edge detection. Step VCircular Hough transform has been applied in order to get iris and pupil area from the image. Step VIInner o0f the pupil area has been excluded from the image as it is not a biometric tool and iris region has been kept as original. Step VIIUpper and lower eyelids have been removed which results in required iris neede for matching. Step VIIIUpper and lower eyelids have been removed which results in required iris needed for matching. Step IXFinally template has been generated and has been matched to the existed database. V. EXPERIMENTAL RESULTS The software implementation of the project has been done using MATLAB. MATLAB stands for MATRIX LABORATORY, software developed by Mathsworks honor in USA. Procedure of the proposed algorithm has been explained as below: It comprises of few basic modules: Image Preprocessing, Phase -1 Iris Localization, Phase-2 Iris Localization, and Fitting Non-circular Contours. Literature reveals that circular hough transform(CHT) is tolerant to broken contours of the objects in the ideal images. However It may not be true for the non ideal data. It may be because of the non uniform illumination, non circular iris contours and occlusions such a hair, glasses, contact lens, eyelids and eyelashes. However to make CHT robust for the non ideal data as well, we augment it by the image gray level, statistics for example, global average gray level intensity, lower and upper saturated gray level limits of the eye image. This combination of CHT and gray level statistics of image results in a better strategy for finding iris. It implies that a circular region in an eye image would be considered as an iris/pupil region, provided the following two conditions are true:  A peak corresponding to the pupil/iris circle should be present in CHT accumulator; and  Gray level intensity of that circular region should be relatively low with respect to some threshold value. Next, module localizes the pupil and iris circles in the preprocessed eye image using an effective scheme. As we mentioned earlier, our focus is on precise localization of the iris inner and outer contours, therefore the final localized iris may contain eyelid and eyelashes occlusions as in. International Journal of Emerging Technology and Advanced Engineering Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 4, Issue 7, July 2014) 961 Results for the above discussed are shown in images as below Figure 1:original image taken for discussion Figure 2: Eye region extracted from the original image Figure 3: Segmentation of extracted region Figure 4: Edge Detection of segmented image Figure 5: Iris and pupil markings Figure 6: Upper and lower eyelid markings Experimental results for two databases taken rom testing is shown below International Journal of Emerging Technology and Advanced Engineering Website: www.ijetae.com (ISSN 2250-2459, ISO 9001:2008 Certified Journal, Volume 4, Issue 7, July 2014) 962 Table 1: Shows results for database 1 as 1 for true and -1 for false in the corresponding column matched or not matched Database1 Matched NotMatched Iris detected