DocumentCode :
1354207
Title :
An Event-Driven Multi-Kernel Convolution Processor Module for Event-Driven Vision Sensors
Author :
Camuñas-Mesa, Luis ; Zamarreño-Ramos, Carlos ; Linares-Barranco, Alejandro ; Acosta-Jiménez, Antonio J. ; Serrano-Gotarredona, Teresa ; Linares-Barranco, Bernabé
Author_Institution :
Dept. of Eng., Univ. of Leicester, Leicester, UK
Volume :
47
Issue :
2
fYear :
2012
Firstpage :
504
Lastpage :
517
Abstract :
Event-Driven vision sensing is a new way of sensing visual reality in a frame-free manner. This is, the vision sensor (camera) is not capturing a sequence of still frames, as in conventional video and computer vision systems. In Event-Driven sensors each pixel autonomously and asynchronously decides when to send its address out. This way, the sensor output is a continuous stream of address events representing reality dynamically continuously and without constraining to frames. In this paper we present an Event-Driven Convolution Module for computing 2D convolutions on such event streams. The Convolution Module has been designed to assemble many of them for building modular and hierarchical Convolutional Neural Networks for robust shape and pose invariant object recognition. The Convolution Module has multi-kernel capability. This is, it will select the convolution kernel depending on the origin of the event. A proof-of-concept test prototype has been fabricated in a 0.35 μm CMOS process and extensive experimental results are provided. The Convolution Processor has also been combined with an Event-Driven Dynamic Vision Sensor (DVS) for high-speed recognition examples. The chip can discriminate propellers rotating at 2 k revolutions per second, detect symbols on a 52 card deck when browsing all cards in 410 ms, or detect and follow the center of a phosphor oscilloscope trace rotating at 5 KHz.
Keywords :
CMOS image sensors; convolution; image sequences; neural nets; 2D convolution module; CMOS process; computer vision system; continuous stream; convolution kernel; event-driven dynamic vision sensor; event-driven multikernel convolution processor module; hierarchical convolutional neural network; high speed recognition; invariant object recognition; multikernel capability; phosphor oscilloscope; visual reality; Arrays; Convolution; Feature extraction; Kernel; Sensor systems; Address-event representation (AER); asynchronous vision sensors and processors; high-speed imaging; image convolutions; image sensors; machine vision; neural networks hardware; neuromorphic circuits; robot vision systems; visual system;
fLanguage :
English
Journal_Title :
Solid-State Circuits, IEEE Journal of
Publisher :
ieee
ISSN :
0018-9200
Type :
jour
DOI :
10.1109/JSSC.2011.2167409
Filename :
6054033
Link To Document :
بازگشت