Detection Methods: part 2

Additional Notes on Methods 1 and 2

The two matched filter approaches produce limited success as applied so far. We are quite successful at detecting an eye for which we have specifically designed a filter, but this obviously will not be a desirable solution in practical applications.

We find that the application of a matched filter designed for one image does generally produce a detection spike when applied to another image, but false peaks on other features are also produced which are often larger than the peaks at the true eyes.

In addition, we find that the matched filter generally detects only one of the two eyes in an image. Specifically, the filter produces the highest peak at the eye for which the filter is matched and produces a lesser peak at the other eye. While the detection of both eyes may usually be accomplished by the selection of a suitable detection threshold, the magnitude of the primary and secondary peaks and the magnitude of the other high points in the image background varies widely from image to image. The automated selection of an appropriate threshold is therefore somewhat difficult.

If we assume that each image will contain exactly two eyes, a dection algorithm may simply find the two highest peaks in the result. Such an assumption is actually warranted for some applications. In facial recognition for security systems, for example, the subject may be directed to face the camera and the lighting conditions may be controlled. The algorithm in such a system need not be made insensitive to problems such as partial obscuration of an eye by hair, an oblique viewing angle, etc. (Allowing such a system to handle humans with fewer than two eyes does present a problem, however.)

Another approach would involve restricting the search region to the area of the face where the eyes are expected to be found. This approach would require only the rough estimation of the boundaries of the face in the image. This approach would require the restriction, however, that each image would contain exactly one face.

Previous Page Next Page

jchen@micro.ti.com
tdorney@ti.com
sparr@owlnet.rice.edu