DocumentCode :
1867266
Title :
Statistical signatures for self-adaptive sensing
Author :
Price, E.I. ; Reece, S. ; Probert-Smith, P.
Author_Institution :
Dept. of Eng. Sci., Oxford Univ., UK
fYear :
2001
fDate :
2001
Firstpage :
141
Lastpage :
146
Abstract :
Designing signal processing software is difficult. it is difficult because the world is unpredictable and it is impossible to guarantee software reliability in unforeseen circumstances. Further, it is difficult to anticipate sensor behaviour, as ambient conditions-for example, lighting or weather, can affect the data they output in a multitude of ways. Autonomous, on-line, self-adaptive image processing software is required, that can be adjusted when novel sensing environments are encountered. The appropriate choice of sensor and signal processing tools is a matter of context and the contextual consensus that is available within the framework of a multiple sensor system.
Keywords :
adaptive systems; image processing; online operation; sensor fusion; statistical analysis; ambient conditions; autonomous online self-adaptive image processing software; contextual consensus; multiple sensor system; self-adaptive sensing; signal processing software design; signal processing tools; statistical signatures; Adaptive signal processing; Design engineering; Process design; Reliability engineering; Robot sensing systems; Sensor phenomena and characterization; Sensor systems; Signal processing; Signal processing algorithms; Software design;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multisensor Fusion and Integration for Intelligent Systems, 2001. MFI 2001. International Conference on
Print_ISBN :
3-00-008260-3
Type :
conf
DOI :
10.1109/MFI.2001.1013522
Filename :
1013522
Link To Document :
بازگشت