Vibratory Image Feature Extraction Based on Local Log-Polar Symmetry

We propose a feature extraction method for a newly developed vibratory image sensor. The proposed method extracts a local log-polar symmetry as image feature. In this article, it is shown that log-polar symmetric patterns include well-known characteristic points such as edges, corners, peaks, and so on. The principle of our method is as follows: a periodic vibration of an image sensor modulates radial and angular components of local image pattern into separated temporal frequency components of incident light on each pixel. Comb type filters after a photo detector decompose the radial component and the angular one, and accumulate their powers in a frame time. These powers are read out for calculating unevenness between them. The unevenness indicates the strength of symmetry of corresponding local image pattern. Because our method modulates local image pattern using image sensor’s vibration, the extracted features are free from spatial quantization errors.