测试科学与仪器2010，Vol.1Issue(2)：116-120,5.DOI:10.3969/j.issn.1674-8042.2010.02.04
Improved Real-time Implementation of Adaptive Gassian Mixture Model-based Object Detection Algorithm for Fixed-point DSP Processors
Improved Real-time Implementation of Adaptive Gassian Mixture Model-based Object Detection Algorithm for Fixed-point DSP Processors
摘要
Abstract
Foreground moving object detection is an important pocess in various computer vision applicatipons such as intelligent visual sur-veillance,HCI,object-based video compression,etc.One of the most successful moving object detection algorithms is based on Adaptive Gaussian Mixture Model (AGMM).Although AGMM-based object detection shows very good performance with respect to object detection accuracy,AGMM is very complex model requiring lots of floating-point arithmetic so that it should pay for expensive computational cost.Thus,direa implementation of the AGMM-based object detec-tion for embedded DSPs without floating-point arithmetic HW support cannot satisfy the real-time processing requirement.This paper pre-sents a navel real-time implementation of adaptive Gaussian mixture model-based moving object detection algorithm for fixed-point DSPs.In the proposed implementation,in addition to changes of data types into fixed- point ones,magnification of the Gaussian distribution tech nique is introduced so that the integer and fixed-point arithmetic can be easily and consistently utilized instead of real number and floating-point arithmetic in processing of AGMM algorithm.Experimental re-sults shows that the proposed implementation have a high potential in real-time applications.关键词
background modeling/real-time computing/object de-tectionKey words
background modeling/real-time computing/object de-tection分类
天文与地球科学引用本文复制引用
Byung-eun LEE,Thanh-binh NGUYEN,Sun-tae CHUNG..Improved Real-time Implementation of Adaptive Gassian Mixture Model-based Object Detection Algorithm for Fixed-point DSP Processors[J].测试科学与仪器,2010,1(2):116-120,5.基金项目
This work was supported by Soongsil University Research Fund and BK21 of Korea （）
