A 0.5 V, 14.28-kframes/s, 96.7-dB Smart Image Sensor With Array-Level Image Signal Processing for IoT Applications.

This paper presents a smart image sensor with multiple operation modes, including edge extraction, multipoint tracking (MPT), and high-dynamic-range (HDR) imaging for wireless sensor nodes in Internet-of-Things applications. The pixel consists of a 0.5 V operated pulsewidth-modulation sensor for achieving HDR response and low fixed pattern noise. Array-level image signal processing is implemented using a local interpixel feedback network, in-pixel low power dynamic logics, and a corresponding peripheral with an event-driven hand-shaking readout. A prototype chip with a $64 \times 64$ CMOS image sensor (CIS) array is designed and fabricated in a TSMC 0.18- \mu \text{m} CIS technology. In the MPT mode, the measured tracking speed is 14.28 kframes/s with an error of 0.36 pixels and a tracking capability up to four points. In the HDR imaging mode, the achieved DR is 96.7 dB. [ABSTRACT FROM PUBLISHER]