Title :
Detecting moving objects from a moving platform
Author :
Frazier, J. ; Nevatia, R.
Author_Institution :
Inst. for Robotics & Intelligent Syst., Univ. of Southern California, Los Angeles, CA, USA
Abstract :
The authors present a system that successfully detects edge segments of moving objects viewed by a translating camera. The system uses two key concepts. First, a complex logarithmic mapping (CLM) is used to simplify detection of the desired movement. This converts the problem from one of detecting a complex motion along both the X and Y axes, to one of detecting vertical motion along an angular axis. Second, a mechanism to detect vertically moving edges has been developed to detect the requisite vertical motion within the CLM image. Results of processing both synthetic and real image sequences are shown. The synthetic sequence consisted of imagery, containing a moving truck, taken from an observation vehicle that was following a road. The system correctly detected portions of the truck as it passed the observer. Important similarities exist between biological vision systems and the techniques presented for motion detection
Keywords :
edge detection; image sequences; motion estimation; biological vision systems; complex logarithmic mapping; complex motion; edge detection; motion detection; motion estimation; moving objects detection; moving platform; moving truck; observation vehicle; real image sequences; synthetic images; translating camera; vertical motion detection; Cameras; Detectors; Image converters; Image edge detection; Image motion analysis; Image sequences; Intelligent robots; Layout; Motion detection; Object detection;
Conference_Titel :
Robotics and Automation, 1992. Proceedings., 1992 IEEE International Conference on
Conference_Location :
Nice
Print_ISBN :
0-8186-2720-4
DOI :
10.1109/ROBOT.1992.220019