A Real-Time Vision System Using Integrated Memory Array Processor Prototype LSI

This study reports on the performance of a Real-Time Vision System (RVS) and its use of an IMAP prototype LSI. This LSI integrates eight 8 bit processors and a 144 Kbit SRAM on a single chip. Processors operate in SIMD at 200 MIPS (25 MHz). The RVS was developed using 64 IMAP prototype LSIs connected in series in a 512 processor system configuration. Memory on the IMAP prototype LSI can be easily accessed from a host workstation through a VME bus. RVS performance is shown in real-time road-image processing and in a neural network simulation, as well as in low level image processing algorithms, such as filtering, histograms, Discrete Cosine Transform, and rotation. The RVS image processing is shown to be much faster than the video rate. INTRODUCTION Figure 2 shows an example of a real-time vision system with 512x512 pixel images. Using four of the IMAPs shown in Fig.1, the system contains 512 processors, each of which carries out image processing to an individual column of images. Since the publication of that study, the authors have developed an IMAP prototype LSI which integrates eight 8 bit processors and a 144 Kbit SRAM, and a Real-Time Vision System (RVS), which uses 64 IMAP prototype LSIs. The purpose of this development has been to investigate IMAP potential, to test sequencer designs, to develop software tools and to develop image processing algorithms. This paper describes the IMAP prototype LSI and the RVS, and it discusses the the results of its performance in image processing and in neural network simulation on the RVS .

[1]  S. S. Wilson,et al.  The AIS-5000 Parallel Processor , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  A. Masaki,et al.  A wafer scale integration neural network utilizing completely digital circuits , 1989, International 1989 Joint Conference on Neural Networks.