计算机科学
图像传感器
帧速率
频道(广播)
视频阵列图形
事件(粒子物理)
实时计算
帧(网络)
图像分辨率
计算机视觉
通用异步收发机
人工智能
计算机硬件
电信
物理
炸薯条
量子力学
现场可编程门阵列
作者
C.G. Jakobson,Rami Fraenkel,Nimrod Ben-Ari,Roman Dobromislin,N. Shiloah,Tomer Argov,Willie Freiman,Gal Zohar,L. Langof,Oren Ofer,Rahel Elishkov,Edan Shunem,Michael Labilov,Menashe Alcheck,Yiftah Kalfa,M. Nitzani,Yoram Karni,I. Shtrichman,Tuvy Markovitz
摘要
Event-based imagers are an emerging class of sensors with demonstrated advantages relative to traditional imagers. Eventbased vision sensors have a limited number of output bits that are only responsive to image variations, thus overcoming the speed and power constraints of the conventional imagers based on image integration. So far, event-based vision has been implemented in visible CMOS sensors. SCD has developed a new event-based VGA/10µm InGaAs sensor that is sensitive in the Visible to SWIR band (600–1700nm) thus extending the standard imaging capabilities such as day light, low-light level, see-through fog and dust. Another novelty of this sensor is that the event channel outputs in parallel to a conventional integrating fast imaging channel. Moreover, the event channel can be reconfigured to provide a fast laser pulse detection mode, which also outputs in parallel to the integrated image. The new ROIC outputs standard video at 13 bit resolution with a high frame rate of 800 frames per second and can double this rate by lowering the resolution to 11 bit. The Event channel can reach up to 25 kHz of negative, null, or positive signal and the laser channel can double this output providing a single bit detection frame up to 50 kHz. In this work, we will elaborate on the architecture, key features, and preliminary simulations of the ROIC and sensor
科研通智能强力驱动
Strongly Powered by AbleSci AI