Bio-inspired design: nonlinear digital pixels for multiple-tier processes

Abstract

The pixel array in a conventional image sensor performs worse than the human retina mainly in dynamic range and dark limit. These limitations may be overcome by introducing others, but we aim to overcome them without limitation given biological precedent and inspiration. Whereas conventional image sensors use linear analog pixels in a single-tier process, we design nonlinear digital pixels for multiple-tier processes. A wide dynamic range is easily achieved with nonlinearity, while image quality is ensured through digital signal processing and in-pixel analog-to-digital conversion. For low dark limit and high spatial resolution, we exploit the high fill factor and heterogeneous integration of emerging multiple-tier processes. Our progress is demonstrated with experimental results from three image sensor prototypes, which provide supporting evidence for the proposed approach.

Digital Pixel Sensor Array with Logarithmic Delta-Sigma Architecture

“…Denoted f SR , the SR is computed as follows, from the focal-plane distance, d , subtended by one scene degree [16]: $f_{\text{SR}}=d\cdot 0.5/p$ $d=2\ell tan(0.5°)$ Here, p is the pixel pitch (38 μm) and 0.5/ p is the spatial Nyquist sampling rate. A more accurate definition requires the modulation transfer function (MTF) [4], which is difficult to compute unless aliasing is incorrectly ignored.…”

“…Sample calculations and a data table, which includes citations to datasheets and publications, are given by Skorka et al [4,16] for designs 1–24, which were chosen based on documentation and diversity. In each scatter plot, designs are included if at least one parameter is known.…”

“…Imaging performance relative to other image sensors, adapted from Skorka et al [4,16]. ( a ) Visual field versus power consumption; ( b ) spatial versus temporal resolution; ( c ) peak SNDR and/or SNR versus dynamic range; and ( d ) dark limit versus dynamic range.…”

“…State-of-the-art image sensors also suffer from high DL, relative to the human eye, but their SR is competitive [4,16]. Fortunately, emerging methods for vertical integration are compatible with the proposed architecture, and these methods are expected to enable lower DL and higher SR [17].…”