Iris recognition using discrete cosine transform and Kekre's fast codebook generation algorithm
Proceedings of the International Conference and Workshop on Emerging Trends in Technology
Proceedings of the International Conference and Workshop on Emerging Trends in Technology
| Hi-index | 0.00 |
Iris recognition is one of the most reliable biometric technologies. In this paper, we presented a novel method for iris recognition, using a complex mapping procedure and best-fitting line for the iris segmentation and 1D Gabor filter with 2DPCA for the recognition approach. We used an intensity threshold method with Canny edge detector to extract the rough region of the pupil. For the outer boundary a median filter with prewitt compass edge detector were used to localize the rough region of the outer boundary. By selecting the bottom point of the pupil (which is not usually occluded by the eyelids and eyelashes) as a reference point, two sets of intersecting points between the horizontal lines and pupil’s inner and outer boundaries were created. Each point set was map into a new complex domain using the complex inversion map function and the best-fitting line was found on the range. Exact inner and outer boundaries of the iris were found by remapping the best-fitting lines to original domain. In the recognition procedure, we used the real term of 1D Gabor filter. In order to reduce the dimensionality of the extracted features, the new introduced 2DPCA method was used. We tested our proposed algorithm by implementing a ground truth method. Experimental results show that the proposed method has an encouraging performance in both segmentation and recognition approaches.