为解决空间碎片漫反射激光测距回波数据信噪比低,难以快速高效地提取有效数据点的问题,提出一种基于随机Hough变换的有效回波提取方法。考虑O-C残差中相邻信号点之间的时间相关性,将有效回波数据的提取问题转化为图像空间中的曲线识别问题。采用随机Hough变换,将确定一条曲线的n个数据点映射为参数空间中的一个点,并在参数空间中运用分数累加的策略进行曲线辨识,提取信号点。采用该方法处理中国科学院云南天文台空间碎片漫反射激光测距实验得到的数据,结果表明,该方法可以快速地提取近似曲线型分布的有效回波,且占用内存少,漏检率低,为自动高效提取空间碎片漫反射激光测距有效回波提供了新思路。 In order to solve the problem of low signal-to-noise ratio (SNR) of space-time diffuse laser ranging echo data and difficulty of extracting effective data points quickly and efficiently, an effective echo extraction method based on random Hough transform is proposed. Considering the temporal correlation between adjacent signal points in O-C residuals, the extraction of effective echo data is transformed into the problem of curve identification in image space. Using random Hough transform, the n data points which define a curve are mapped to a point in the parameter space, and the curve is identified by using the strategy of fractional accumulation in the parameter space to extract signal points. The results show that this method can extract effective echoes of approximate curve-type distribution quickly with less memory and low detection rate, and is an automatic It provides a new idea to efficiently extract effective echo of space reflection diffuse laser ranging.