基于ARM和DSP的双核嵌入式视频监控系统  被引量:4

Dual-core Embedded Video Monitoring System Based on ARM and DSP

在线阅读下载全文

作  者:张磊[1] 卢刚[1] 彭力[1] 

机构地区:[1]江南大学物联网工程学院,江苏无锡214122

出  处:《计算机测量与控制》2017年第6期49-52,67,共5页Computer Measurement &Control

摘  要:针对传统嵌入式视频监控设备集成度低,功能有限,响应速度慢,计算能力不足的问题,研究了一种基于ARM微处理器S3C6410与TMS320DM642DSP双核体系结构的嵌入式视频监控系统;该系统采用CMOS摄像头采集视频数据,经DM642上算法处理后通过HPI高速并行接口传输给ARM处理器,ARM处理器负责对接收到的视频进行控制和LCD显示;重点阐述了ARM和DSP系统单元间的硬件连接方案,HPI接口驱动程序的设计实现,以及结合Linux异步通知和多线程技术的QT/Embedded客户端人机交互软件的编写;最后,移植了运动目标跟踪算法到DSP中并对整个系统进行实验测试,验证系统的性能;经实验证明,所设计的系统稳定可靠,视频清晰流畅,具有性能高、功耗和成本低、可扩展性强等优点,在嵌入式视频监控领域具有推广和使用价值。Aiming at the low integration, limited function, slow response and insufficient computing power of traditional embedded video surveillance application, this paper studied an embedded video monitoring system on the basis of dual--core structure of ARM microprocessor $3C6410 and TMS320DM642 DSP. The system uses CMOS camera to collect video data, which will be transmitted to the ARM processor through the high--speed parallel interface after the algorithm processing by DM642. The ARM processor is responsible for controlling the re- ceived video and displaying on LCD. This paper emphasizes on the hardware connection scheme among system units of ARM and DSP, the design and implementation of HPI driver, and the software programming of QT/Embedded application combining Linux asynchronous notifi- cation and multi thread technology. Finally, transplanted target tracking algorithm into the DSP and tested the whole system to verify the performance of the system. Experiments results prove that the system is stable and reliable, the video is clear and smooth, with high per- formance, low power consumption, low cost, scalability and other advantages. It will has more value of promotion and using in the field of embedded video monitoring.

关 键 词:S3C6410 DSP QT 视频监控 

分 类 号:TP277[自动化与计算机技术—检测技术与自动化装置]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象