Title :
The evolution of a massively parallel vision system for real-time automotive image processing
Author_Institution :
Dipartimento di Ingegneria dell´´Inf., Parma Univ., Italy
Abstract :
This paper presents the first prototype of the PAPRICA massively parallel system which is integrated in the MOB-LAB experimental land vehicle for real-time vision-based road marking detection. Its main bottlenecks are highlighted and its evolution toward a linear array is discussed. This system has been enhanced with a simple but powerful interprocessor communication network for the exchange of information among processors not directly connected, which allows an extremely efficient implementation of the road marking detection algorithm as well as other morphological applications
Keywords :
automated highways; computer vision; image recognition; multiprocessor interconnection networks; parallel architectures; real-time systems; road vehicles; traffic engineering computing; MOB-LAB experimental land vehicle; PAPRICA; interprocessor communication network; linear array; massively parallel vision system; morphological applications; prototype; real-time automotive image processing; vision-based road marking detection; Automotive engineering; Communication networks; Detection algorithms; Image processing; Land vehicles; Machine vision; Pixel; Prototypes; Real time systems; Roads; Vehicle detection;
Conference_Titel :
Parallel Processing Symposium, 1996., Proceedings of IPPS '96, The 10th International
Conference_Location :
Honolulu, HI
Print_ISBN :
0-8186-7255-2
DOI :
10.1109/IPPS.1996.508138
