Spatiotemporal alignment of sequential images: Utilization of GPS data as reference to alignment of multiple video cameras

Autor:	Walter C. S. S. Simoes, Diego S. Froner, José Luiz de Souza Pio
Rok vydání:	2016
Předmět:	business.industry Real-time computing Data matching Tracking (particle physics) GPS signals Object (computer science) Geography Position (vector) Gps data Global Positioning System Computer vision Artificial intelligence Degree of certainty business
Zdroj:	CLEI
DOI:	10.1109/clei.2016.7833351
Popis:	Many applications developed for tracking moving objects apply multiple cameras distributed in the study scenario. However, this form requires a lot of effort to know the time position of the moving object to align the frames at the same time and position. This paper presents a study on the camera space-time alignment to increase the degree of certainty about the location of objects monitored by merging information obtained from cameras and GPS. The results obtained by testing on simulator and a study scenario showed a degree of 98% and 71.42% accuracy. The lower result in the actual scenario was due to GPS signal use limitation. The major contribution of this work is to provide an alternative to future works which propose solutions to the problem of spatiotemporal alignment of cameras without overlapping fields.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::0e91f2601b0e86a969d061fe4c501317 https://doi.org/10.1109/clei.2016.7833351 Zobrazit plný text záznamu