Motion detection in moving pictures
Yükleniyor...
Tarih
2005
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Motion analysis is one of the major techniques which is used in applications related with interpretation, processing and compressing digital images. In this thesis, seven different methods that are widely used for the motion detection in a moving picture sequence, were studied and implemented, and their performances were discussed. These methods were investigated under two major headings namely Background Subtraction Methods and Statistical Methods. The performance tests for the methods were carried out for the images that were taken under various illumination conditions. The results showed that the background subtraction methods are very sensitive to sudden illumination changes on a running sequence which is a major drawback. While both of the methods are quite satisfactory for indoor scenes that provide stable illumination conditions, the methods that were studied under secondary group of the statistical methods have been found to be successful for the outdoor scenes that present sudden illumination changes. Although, the pitfalls of the methods resulting sudden illumination changes have been improved, the need for taking the images under a carefully illuminated environment, needs further investigation. © 2005 IEEE.
Açıklama
IEEE 13th Signal Processing and Communications Applications Conference, SIU 2005 -- 16 May 2005 through 18 May 2005 -- Kayseri -- 69003
Anahtar Kelimeler
Image compression, Lighting, Motion pictures, Statistical methods, Background Subtraction Methods, Digital images, Motion detection, Moving picture sequences, Motion estimation
Kaynak
Proceedings of the IEEE 13th Signal Processing and Communications Applications Conference, SIU 2005
WoS Q Değeri
Scopus Q Değeri
N/A
Cilt
2005
