Title :
Extracting Local Binary Patterns with MIPA4k vision processor
Author :
Lahdenoja, O. ; Poikonen, J. ; Laiho, M.
Abstract :
In this paper we show how the MIPA4k vision processor can be used to extract and process local image features called Local Binary Patterns (LBPs). Using these low level image features more advanced higher level processing can be performed. LBP histograms can be extracted with wide field of potential applications. Also, directional thresholding results, which are part of LBP extraction process, can be used for higher level processing such as face localization. The MIPA4k vision processor consists of an array of 64 ?? 64 mixed-mode processing units with on-chip image capture sensors.
Keywords :
feature extraction; image processing; image sensors; MIPA4k vision processor; directional thresholding; face localization; higher level processing; image features; local binary patterns; on-chip image capture sensors; Application software; Cellular networks; Data mining; Equations; Face detection; Histograms; Information technology; Pattern analysis; Pixel; Sensor arrays;
Conference_Titel :
Cellular Nanoscale Networks and Their Applications (CNNA), 2010 12th International Workshop on
Conference_Location :
Berkeley, CA
Print_ISBN :
978-1-4244-6679-5
DOI :
10.1109/CNNA.2010.5430265