论文部分内容阅读
本文提出了一种序列分级搜索二值图象相关检测的电视跟踪方法,它综合了SSDA法和序列分级搜索法的优点,可以简化算法,将多比特的相关检测变成二值图象的相关检测,使实现处理的硬件规模减小,运算速度提高;采用平滑滤波改善图象信噪比,可以提高二值图象相关检测的性能,用并行运算实现相关检测的实时处理。本文对二值图象相关检测的性能作了分析和计算机模拟。结果表明,当图象信噪比为1、几何失真因子为10％时,该方法仍有较好的性能。这种方法不仅可用于电视跟踪,也可以用于二维成象的雷达、红外跟踪系统。
In this paper, we propose a TV-tracking method for sequence-level binary image correlation detection, which combines the advantages of SSDA and sequence-level search to simplify the algorithm and transform multi-bit correlation detection into binary image correlation So that the size of the hardware to be processed can be reduced and the computing speed can be improved. The smoothing filter is adopted to improve the signal-to-noise ratio of the image, which can improve the performance of the binary image correlation detection and realize the real-time processing of the relevant detection by parallel computing. This paper analyzes the performance of binary image correlation detection and computer simulation. The results show that the proposed method still has better performance when the image signal to noise ratio is 1 and the geometric distortion factor is 10%. This method can not only be used for television tracking, can also be used for two-dimensional imaging of the radar, infrared tracking system.