The use of the Hough Transform Essay

The parabolic Hough transform controls the curvature, which is the peak of parabola and it is the angle of rotation relative to the xaxis. According to reference [6], in edge detection step, we take the partial derivatives with respect to the x axis (horizontal direction) for detecting the outline of eyelids whereas again we take the partial derivatives with respect to the y axis (vertical direction). It is used for detecting the outer circular boundary of the iris. Reason behind this is, the eyelids are usually positioned in horizontal alignment, and also the eyelid edge maps will imprecise the circular boundary of iris edge map if using all incline data. Only the vertical pitch are considered for the iris boundary location and it will diminish influence of the eyelids when performing circular Hough transform, and all of the edge pixels defining the circle are not required for successful localization. These make more efficient and accurate circle localization because there are few edge points to cast votes in the Hough space.

IV. SEGMENTATION

This segmentation (localization) process is to search for the center coordinates of the pupil and the iris along with their radius. These coordinates are marked as ci,cp where ci represented as the parameters [xc, yc, r] of the limbic and iris boundary and cp represented as the parameters [xc, yc, r]of the pupil boundary. It makes use of [6] to select the possible center coordinates first. The method consist of threshold followed by checking if the selected points (by threshold) correspond to a local minimum in their immediate neighborhood these points serve as the possible center coordinates for the iris.

These radius values were set manually the input for this function is the image to be segmented and the input parameters in this function including rmin and rmax (the minimum and maximum values of the iris radius). The range of radius values to search for was set manually, depending on the database used. The output of this function will be the value of ci and cp which is the value of [xc, yc, r] for the pupilary boundary and the limbic/iris boundary and also the segmented image. For all the above process the use of the Hough Transform is made as discussed in the above section. However, the Segmentation process was not able to perfectly reconstruct the same pattern from images with varying amounts of pupil dilation, as the deformation of an iris results in small changes of its surface patterns.

V. NORMALIZATION

After the process of iris segmentation, the next step is normalization. From the process of normalization, the segmented image of the eye will give the value radius pupil and the iris. This image will be cropped base on the value of iris radius, so that the unwanted area will be removed (e.g. sclera and limbic),arcus senilis is described as a yellowish white ring around the cornea that is separated from the limbus by a clear zone 0.3 to 1 mm in width. Normally the area