Abstract
With the development of current information technology, health diagnosis based on iris analysis and biometrics has received more attention. Iris localisation is an important phase in iris recognition. Iris localisation is not an easy task, and dealing with non-ideal iris images could cause an incorrect location in the iris localisation. The conventional methods for iris location involve many searches, which can be noisy and out-dated. These techniques could be inaccurate while describing the pupillary boundaries and also could lead to many errors while carrying out feature recognition and extraction. Hence, for addressing this issue, this paper proposes a method for iris localisation in the case of ideal and non-ideal iris images. In this study, the algorithm is based on finding the region of interest (ROI) classification with the help of a Support Vector Machine (SVM) and applying a histogram of grey level as a descriptor in each region from the region growing. The valid ROI found from the probabilities graph of the SVM was obtained by looking at the global minimum conditions determined by a second derivative model in a graph of functions. This helps in elimination of the sensitive noises and decreasing the calculations while reserving relevant information as far as possible. Subsequently, the classified image will be localised by using the Hough Transform method. The experimental results presented in this study indicate that the proposed algorithm efficiently improved the Hough Transform method in localising the boundary of the iris.