DocumentCode :
2975544
Title :
Point process texture generation model: implications to multi-resolution analysis
Author :
Zacksenhouse, Miriam ; Konstantino, Eitan
Author_Institution :
Fac. of Mech. Eng., Technion-Israel Inst. of Technol., Haifa, Israel
fYear :
1999
fDate :
1999
Firstpage :
195
Lastpage :
198
Abstract :
Textures are generated and characterized using point processes whose realizations localize the (boundaries of) texture primitives. For simplicity, one dimensional point processes are used to generate parallel independent cross sections of the image. The interval statistics of the underlying point process describes the spatial parameters of the texture primitives in the studied orientation. Multiple resolution texture analysis methods are analyzed in the context of the proposed texture generation model and related to the point process interval statistics. A variant of these methods is proposed to directly estimate the interval statistics, and thus the spatial parameters of the texture primitives
Keywords :
image resolution; image texture; parameter estimation; statistical analysis; interval statistics; multi-resolution analysis; multiple resolution texture analysis; one dimensional point processes; parallel independent cross sections; point process interval statistics; point process texture generation model; spatial parameters; texture primitives; Character generation; Context modeling; Image texture analysis; Spatial resolution; Statistical analysis; Statistics;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Higher-Order Statistics, 1999. Proceedings of the IEEE Signal Processing Workshop on
Conference_Location :
Caesarea
Print_ISBN :
0-7695-0140-0
Type :
conf
DOI :
10.1109/HOST.1999.778724
Filename :
778724
Link To Document :
بازگشت