摘要
针对传统的聚焦评价函数运算量大、峰值搜索算法易受干扰且视频质量反复变化的问题,实现一种较传统聚焦评价函数与爬山算法相结合的聚焦算法更好的自动聚焦算法。采用TI公司的TMS320DM368处理器自带的自动聚焦(AF)引擎作为硬件实现的聚焦评价函数;采用的峰值搜索算法将聚焦曲线分为聚焦区和散焦区,在散焦区使用大步长的GM(1,1)模型预测搜索方向,在聚焦区使用小步长的爬山算法搜索峰值。实验结果表明,该算法拥有更好地实时性与准确性,适用于高清网络摄像机的自动聚焦。
To solve the problems of the traditional focusing evaluation function including the high computational load, the susceptibility to interference of peak searching algorithms and the change of the video quality, a better automatic focusing method was realized. The TMS320DM368 processor with the AF engine of the TI was adopted as the focusing evaluation function, which was implemented by the hardware. The peak searching algorithm divided the autofocus curve into the focus area and the defoeused area. The GM（1, 1） model was used to predict the search direction with big steps in the defocused area, while mountain climbing algorithm was used to search the peak with small steps in the focus area. The experimental results show that the algorithm is real-time and accurate and is suitable for the high-definition network camera.
出处
《计算机工程与设计》
CSCD
北大核心
2014年第9期3289-3294,共6页
Computer Engineering and Design
