Dynamic modulation transfer function of a retina-like sensor

Abstract

In this paper, we propose a method to deduce the dynamic modulation transfer function (DMTF) of a space-variant sampling retina-like sensor and demonstrate its utilization in the forward motion imaging process. With the analysis of sampling and the motion imaging property of the sensor, DMTF has been derived. Next, the performance of DMTF between a retina-like sensor and a rectilinear sensor is compared, and the results show that the degradation of DMTF in forward motion is less than that of a rectilinear sensor. Then, the output images are obtained through simulation based on DMTF, and they are compared with that obtained from a CMOS camera with the same forward motion conditions. The Pearson correlation coefficients between the two kinds of images are all larger than 0.85. Thus, the effectiveness of DMTF is shown.

Modeling and simulations on retina-like sensors based on curved surface

“…Bio-inspired image sensors based on a retina-like structure have been attracting researchers for many years [1][2][3][4]. The advantages of retina-like sensors, including multi-resolution sampling [5], log-polar transformation (LPT) from retina to visual cortex [6,7], and invariance of scaling and rotation [8], etc., are beneficial to significantly compress redundant information and track objects in the large field of view (FOV) with high speeds [9,10].…”