Iris recognition is one of effective methods in person identification. Iris locating is the key subject in the process of iris recognition. This paper analyses some limitations of Hough iris locating algorithm, proposes a minimum variance method based on the wavelet analysis to locate the inner boundaries of iris, and uses an improved brightness gradient method to locate the outer boundaries of iris. The result is that the new inner locating method makes good use of the advantages of wavelet analysis, not only simplify the locating algorithm but improve the speed of locating. The new outer locating algorithm decreases the search range effectively and improves the search precision. The experiments of this new iris locating on common iris image show that the minimum variance method based on the wavelet analysis is also get good results.
展开▼