← 返回 JSSC 论文列表JSSC 2006第6期Digital CircuitsNeural Network Accelerator
A Micropower Centroiding
提出一种生物启发的混合视觉芯片，用于实时对象处理，如质心定位、大小测量和计数。
48x48像素分辨率，85μm x 85μm像素尺寸，243μW功耗，1.8V供电，724.64 MIPS/mW计算效率，500μs处理时间
生物启发混合视觉芯片质心定位微功耗并行处理
▸首个能够并行质心定位和大小测量的硅视网膜
▸基于分布式算法的数据驱动脉冲动作
▸高计算效率的微功耗设计
Abstract
A biologically-inspired hybrid vision chip is pre- sented for real-time object-based processing for tasks such as centroiding, sizing and counting of enclosed objects. This system presents the first silicon retina capable of centroiding and sizing multiple objects in true parallel fashion. Based on a novel dis- tributed algorithm, this approach uses the input image to enclose a feedback loop to realize a data-driven pulsating action. The sensor provides a resolution of 48 48 pixels with a 85 m 85 m pixel footprint and has been measured to consume 243 Wa t 1.8-V supply, achieving an equivalent computational efficiency of 724.64 MIPS/mW with a 500- s process time.