Local pixel wise skin detection algorithm based on FPGA

Author

Jin Zhang ; Zaijiao Zhuang ; Weidong Geng

Author_Institution

Key Lab. of Photo-Electron., Nankai Univ., Tianjin, China

fYear

2014

Firstpage

262

Lastpage

265

Abstract

This paper presents an approach to detect skin with taking 16 by 16 pixels as the unit for videos that use Local Pixel Wise Skin Detection algorithm, named LPSD. It reliably recognizes and marks skin for each 16 by 16 pixels in the video, based on Field-Programmable Gate Array(FPGA). A special memory and two special counters reduces storage resource utilization caused by the restore all pixels information in 16 by 16 area. The algorithm could mark the skin area pixel by pixel without restoring 16 by 16 pixels each time. For the VGA standard image, 640 by 480 pixels, this algorithm uses 167,744 memory bits totally.

Keywords

feature extraction; field programmable gate arrays; image reconstruction; image resolution; image segmentation; image sensors; skin; 16 by 16 pixels; FPGA; LPSD; field-programmable gate array; local pixel wise skin detection algorithm; skin marks; storage resource utilization; Cameras; Field programmable gate arrays; Image color analysis; Laboratories; Radiation detectors; SDRAM; Skin; FPGA; memoery resource; skin detection; window process;

fLanguage

English

Publisher

ieee

Conference_Titel

Information Technology and Artificial Intelligence Conference (ITAIC), 2014 IEEE 7th Joint International

Print_ISBN

978-1-4799-4420-0

Type

conf

DOI

10.1109/ITAIC.2014.7065047

Filename

7065047