Iris preprocessing algorithm based on space distance circle marking edge detection

Abstract

The present invention provides an iris preprocessing algorithm based on space distance circle marking edge detection. The method comprises a first step of performing image acquisition on an iris of a human eye through a photographing device, thus to obtain an original gray level image; a second step of carrying out noise processing on the acquired image by using a median filtering algorithm; a third step of decomposing the iris image into four subband images by utilizing Haar wavelets; a fourth step of carrying out edge detection on an inner edge of the iris by using Canny operators; a fifth step of carrying out precision positioning of an inner circuit of the iris by using a sub pixel circle positioning algorithm; and a sixth step of carrying out accurate positioning on an outer circuit by using a method of increasing searching radius step length. According to the iris preprocessing algorithm based on the space distance circle marking edge detection, the Canny operators are utilized for filtering, enhancement and detection, the sub pixel circle positioning algorithm is utilized for carrying out accurate positioning of the inner circle of the iris, and the efficiency of circle detection operators is promoted by using the method of increasing the searching radius step length, thereby realizing accurate positioning of the outer circle of the iris, raising the detection precision, and promoting the detection speed.

Description

technical field [0001] The invention relates to an iris preprocessing algorithm based on space distance circle mark edge detection. Background technique [0002] At present, iris identification is an emerging identification technology in recent years. Its research and application are in the ascendant, and the market provides a broad stage for the development of this technology. The iris pattern is extremely thin and cannot be forged. Compared with automatic fingerprint identification, it has more advantages and is a new technology worthy of research and promotion. Compared with other biometrics, iris has high uniqueness, high precision, and good stability, and is applicable to a wide range of people. [0003] At present, iris recognition is generally through the collection of iris images first, then iris preprocessing, and then iris features are extracted, and finally feature matching is performed with the sample iris feature database extracted from iris samples, and then ...

AI Technical Summary

Problems solved by technology

In this algorithm, the accuracy of the least squares method to determine the inner circle is not high, and the speed of the circle detection algorithm to determine the outer circle is relatively slow, so this algorithm needs to be improved; 4. Use the Sobel operator for edge detection, using zero to second order moments Perform inner circle positioning, and then use the Dougman circle detection operator to perform outer circle positioning. Technically, the Sobel operator is a discrete difference operator, which is used to calculate the approximate value of the gradient of the image brightness function. The disadvantage is that the Sobel operator does not have Strictly simulate the visual physiological characteristics of human beings, so the extracted image contours are sometimes unsatisfactory, and the zero-to-second-order moment and Dougman circle detection operators can only be roughly calculated, so the accuracy cannot meet the expected requirements

Images