[1]周春晖,朱建清,蔡灿辉.采用DaVinci技术的智能监控系统设计与实现[J].华侨大学学报(自然科学版),2011,32(5):506-510.[doi:10.11830/ISSN.1000-5013.2011.05.0506]
　ZHOU Chun-hui,ZHU Jian-qing,CAI Can-hui.Design and Implementation of the Intelligent Surveillance System Based on DaVinci[J].Journal of Huaqiao University(Natural Science),2011,32(5):506-510.[doi:10.11830/ISSN.1000-5013.2011.05.0506]
点击复制

采用DaVinci技术的智能监控系统设计与实现()
分享到：

《华侨大学学报（自然科学版）》[ISSN:1000-5013/CN:35-1079/N]

卷:
第32卷
期数:
2011年第5期
页码:
506-510
栏目:
出版日期:
2011-09-20

文章信息/Info

Title:
Design and Implementation of the Intelligent Surveillance System Based on DaVinci
文章编号:
1000-5013(2011)05-0506-05
作者:
周春晖; 朱建清; 蔡灿辉
华侨大学信息科学与工程学院
Author(s):
ZHOU Chun-hui;  ZHU Jian-qing;  CAI Can-hui
College of Information Science and Engineering, Huaqiao University, Quanzhou 362021, China
关键词:
智能监控系统;  人脸检测;  人脸跟踪;  DaVinci技术;  TMS320DM6446
Keywords:
intelligent surveillance;  face detection;  face tracking;  DaVinci technology;  TMS320DM6446
分类号:
TP277
DOI:
10.11830/ISSN.1000-5013.2011.05.0506
文献标志码:
A
摘要:
采用DaVinci技术,设计一种具有人脸检测与跟踪功能的智能监控系统.利用TMS320DM6446芯片的双核结构优势,在ARM端实现基于Linux系统的多线程应用程序,负责视频录放、用户界面控制、视频信号的网络传输、外围器件控制等; 而在DSP(数字信号处理)端完成基于AdaBoost的人脸检测、基于卡尔曼滤波器和Mean Shift相结合的人脸跟踪及H.264视频编码.实验结果表明:智能监控系统能实现人脸定位与跟踪,准确、实时性高,而且可由具有面向网络数据传输的客户端程序实行远程监控.
Abstract:
An intelligent surveillance system with the features of face detection and tracking based on DaVinci technology is proposed.To exploit the resources of dual-core structure of DM6446,multiple applications,including video recording and playback,user interface control,video signals transmitted over the network,the external device control and so on run in the ARM-side,and those high computational complexity routines,such as AdaBoost-based face detection,face tracking algorithm using Kalman filter combined with Mean-Shift algorithm,and H.264 video codec,are executed in DSP-side.Experiment results show that the system can meet demand of real-time video monitoring,and it can accurately detect and trace the faces,and the data can be transmitted over the network with the client program for the implementation of remote survelliance.

参考文献/References:

[1] 陈晓, 张学杰. 基于DaVinci双核芯片的网络视频服务器的设计技术研究 [J]. 云南大学学报（自然科学版）, 2009, (z1):122-127.
[2] 丁锐, 袁誉乐, 赵勇. 基于达芬奇平台的视频监控系统的设计 [J]. 微计算机信息, 2007, (11/12):44-45.doi:10.3969/j.issn.1008-0570.2007.32.018.
[3] 侯建华, 蓝发财. 基于DaVinci DM6446处理器的DVS的设计与实现 [J]. 中南民族大学学报(自然学科版), 2009(1):72-74.doi:10.3969/j.issn.1672-4321.2009.01.018.
[4] 鲁达, 张文军, 杨华. 基于达芬奇平台的智能视频监控算法设计 [J]. 嵌入式系统应用, 2009(2):5-7.
[5] TI. Codec engine algorithm creator user’s guide [DB/OL]. http://focus.ti.com/lit/ug/sprued6c/sprued6c.pdf, 2007.
[6] TI. Codec engine application developer’s guide [DB/OL]. http://focus.ti.com/lit/ug/sprue67d/sprue67d.pdf, 2007.
[7] TI. TVP5146 data sheet [M]. Chicago:Texas Instruments Incorporated, 2007.
[8] TI. TMS320DM644x DMSoC ARM subsystem reference guide [DB/OL]. http://focus.ti.com/lit/ug/sprue14c/sprue14c.pdf, 2007.
[9] TI. TMS320DM644x DMSoC video processing front end (VPFE) user’s guide [DB/OL]. http://focus.ti.com/lit/ug/sprue38c/sprue38c.pdf, 2007.
[10] Intel. LXT971A single-port 10/100 Mbps PHY transceiver datasheet [DB/OL]. http://www.michael x0.ath.cx/static/projects/sbx2tips/datasheet/Intel/LXT971A.pdf, 2005.
[11] 骆张强. 基于Davinc技术的智能家庭 [D]. 成都电子科技大学, 2007.
[12] TI. TMS320DM644x DMSoC EMAC/MDIO module user’s guide [M]. Chicago:Texas Instruments Incorporated, 2007.
[13] 徐显日, 戴在平, 蔡灿辉. 新型人脸检测算法在视频监控系统中的应用 [J]. 电视技术, 2008(4):89-90.doi:10.3969/j.issn.1002-8692.2008.04.035.
[14] 张涛, 蔡灿辉. 一种改进的Mean Shift实时多人脸跟踪算法 [J]. 计算机应用, 2009(3):781-784.

备注/Memo

备注/Memo:
国家自然科学基金资助项目(60772164)
更新日期/Last Update: 2014-03-23