skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: The MIVS (Modular Integrated Video System) Image Processing System (MIPS) for assisting in the optical surveillance data review process

Conference ·
OSTI ID:6752569

The MIVS (Modular Integrated Video System) Image Processing System (MIPS) is designed to review MIVS surveillance data automatically and identify IAEA defined objects of safeguards interest. To achieve this, MIPS uses both digital image processing and neural network techniques to detect objects of safeguards interest in an image and assist an inspector in the review of the MIVS video tapes. MIPS must be trained'' i.e., given example images showing the objects that it must recognize, for each different facility. Image processing techniques are used to first identify significantly changed areas of the image. A neural network is then used to determine if the image contains the important object(s). The MIPS algorithms have demonstrated the capability to detect when a spent fuel shipping cask is present in an image after MIPS is properly trained to detect the cask. The algorithms have also demonstrated the ability to reject uninteresting background activities such as people and crane movement. When MIPS detects an important object, the corresponding image is stored to another media and later replayed for the inspector to review. The MIPS algorithms are being implemented in commercially available hardware: an image processing subsystem and an 80386 Personal Computer. MIPS will have a high-level easy-to-use system interface to allow inspectors to train MIPS on MIVS data from different facilities and on various safeguards significant objects. This paper describes the MIPS algorithms, hardware implementation, and system configuration. 3 refs., 10 figs.

Research Organization:
Sandia National Labs., Albuquerque, NM (USA)
Sponsoring Organization:
DOE/DP
DOE Contract Number:
AC04-76DP00789
OSTI ID:
6752569
Report Number(s):
SAND-90-0667C; CONF-9007106-31; ON: DE90014326; TRN: 90-025126
Resource Relation:
Conference: Institute of nuclear materials management conference, Los Angeles, CA (USA), 15-18 Jul 1990
Country of Publication:
United States
Language:
English