Image feature detection based on phase congruency by Monogenic filters

The image brightness and contrast changes remain invariant when phase congruency changes. Specifically the monogenic filters, a new image feature detection method is proposed, which is based on phase congruency and the monogenic signal theory in this paper. The performances of Monogenic filters are excelled to that of Log-Gabor filters in theory, hence a greater amount of feature vectors are generated. Compared with the Log-Gabor filters, the most important advantage of Monogenic filters is that it performances with lower time and smaller memory space. The experimental result indicates that image features with Monogenic filters can not only overcome the limitations of Log-Gabor filters, but also improve the location accuracy and anti-noise ability with comparable or better performance.