论文部分内容阅读
研究和分析了激光雷达目标回波与杂波背景的分形特性,以分数布朗运动(FBM)数学模型为基础,通过对不同信噪比(SNR)激光雷达回波的分析,初步证明了激光雷达回波具有布朗运动的特征。针对激光雷达杂波数据具有布朗运动的增量统计自相似性,激光雷达杂波能与分数布朗运动较好地匹配,因此可以采用布朗运动模型对激光雷达回波数据进行分析处理。当信噪比比较低时,杂波和含有目标的回波信号的分形维数比较接近,发生重叠,单靠单一的分形维数无法检测出目标。针对这一问题,分析杂波和目标信号在不同尺度上的分形维数,提出基于不同尺度分形维数的变化特征进行目标检测的算法,即基于多尺度分形维数的目标检测算法。理论分析和实验结果表明该方法具有较高的可靠性和准确性。
Based on the mathematical model of Fractional Brownian Motion (FBM) and the analysis of different SNR (Signal-to-Noise Ratio) lidar echoes, the laser radar Echoes have Brownian motion characteristics. For the clutter data of Lidar, there is an incremental statistical self-similarity of Brownian motion. Lidar clutter can match well with fractional Brownian motion. Therefore, the Brownian motion model can be used to analyze and process Lidar echo data. When the signal to noise ratio is low, the clutter and the echo signal containing the target fractal dimension closer to the overlap, the single fractal dimension alone can not detect the target. To solve this problem, the clutter and the fractal dimension of the target signal at different scales are analyzed, and the target detection algorithm based on the fractal dimension of different scales is proposed, which is the target detection algorithm based on multi-scale fractal dimension. Theoretical analysis and experimental results show that the method has high reliability and accuracy.