在智能电子监控行业,达芬奇双核(ARM+DSP)处理器芯片在视频监控领域扮演了重要角色,依达芬奇技术为依托提出了基于DM6446下的运动目标检测系统;ARM核为多线程设计,DSP核则进行目标检测算法处理;引入的算法在高斯模型统计理论条件下采用背景差分法和帧间差分法相结合的方式对图像进行预处理、相与运算和二值化处理;两者算法相互结合,结合了优点,摒弃了不足;实验结果表明,在基于高端的DM6446平台下,软硬协同良好,能够满足对于光照,背景内容变化不敏感的条件下,实时采集和图像处理的要求。
Davinci dual core (ARM+DSP) processor chip plays an important role in the field of Intelligent electronic monitoring, moving target detection system based on DM6446 was proposed according to Davinci technology. ARM core was used for multi-- threaded design and DSP is used in target detection algorithm processing, the algorithm has been introduced for image preprocessing, AND--ing and binarization processing by combining the two methods of background difference and interframe difference under the statistical theoretical condition of Gaussian model, which Integrated the advantages and Discard the shortages, The experimental results demonstrate under the DM6446 based on high--end platform, hardware and software cooperate well, be able to satisfy the request of real--time acquisition and image processing in the condition which is not sensitive to light, backgroup content changes.