Font Size: a A A

Study On Algorithm Of Iris Location And Recognition

Posted on:2010-06-19Degree:MasterType:Thesis
Country:ChinaCandidate:J N WangFull Text:PDF
GTID:2178360278481428Subject:Signal and Information Processing
Abstract/Summary:PDF Full Text Request
With the development of information technology in contemporary society, the demands for security are becoming more and more important. As the technology of biometrics identification has great advantages, it has been improved and applied in many fields. Iris has very special structures comparing with other numerous biological characteristic, which provides the iris identification with excelsior accuracy. Iris identification technology is regarded as one of the most promising biometric cognition technology.A typical iris recognition system based on iris patterns is composed of iris image acquisition, image preprocessing, feature extraction and coding, matching and recognition. It involves numerous subject domains, such as computer vision and digital image processing, wavelet theory, pattern recognition etc. Iris location and recognition are the key of iris recognition system.This paper locating iris area based on gray characteristics first. This method only needs to search three points on the edge of a circle which were not on the same line to locate the boundary. As the gray changing is not obvious when locates outer boundary, it proposes to find four outer boundary points to locate through two times and chooses a better result as the locating result. Experimental result is shown to illustrate this method has a higher accuracy and precision of location than one time to locate. The correct recognition rate of this method is 98.15%, higher 4.64% than one time location.Then, it takes some necessary image preprocessing. It includes the translation to normalized iris image, bilinear interpolation and image enhancement. In order to reduce the influence caused by parallel moving, size difference and zoom to recognition result, it normalized the iris image. It uses bilinear interpolation to get a logical gray value during normalized the iris image which will get a clear texture. In the end, it enhances the normalized image which is useful for the matching process.It improved the original algorithm which is called iris recognition method based on gray surface matching. It proposed a new method to overcome illumination intensity difference influences the recognition result which occurs during collecting session in this paper. The problem is not effective solved in original algorithm. This algorithm proposes to match after moving two matching feature surfaces to one equivalence gray plane. In order to reduce calculation quantity and enhance the difference between two different kinds iris, this paper computes the square of gray difference surface. The result demonstrates this algorithm is better than original algorithm in correct recognition rate and consumed time. This method's correct recognition rate is 91.61%, higher 3.58% than original algorithm. The original algorithm's average consumed time is 3.42 multiples of this algorithm's, which the average consumed time is 298ms.At last, in order to compare this algorithm with traditional algorithm, it uses phase encoding for iris texture in this paper and finished programming and performance testing of this algorithm.All of these algorithms are tested by the CASIA 1.0 iris database which is supplied by Institute of Automation Chinese Academy of Sciences. This method is proven to suit in practice by the experiment results.
Keywords/Search Tags:Iris recognition, Iris location, Preprocessing, Surface matching, Equivalence gray plane, Phase encoding
PDF Full Text Request
Related items