论文部分内容阅读
在智能激光夜视监控系统对运动目标检测过程中,受到的色噪声干扰较大,导致对运动目标的检测性能下降。传统方法采用分数阶傅里叶变换检测方法,在处理激光夜视监控的非平稳运动目标检测中会产生较大的虚警,检测概率不高。提出一种基于四阶累积量切片后置处理的改进的分数阶傅里叶变换检测算法实现对智能激光夜视监控系统中的运动目标检测。进行智能激光夜视监控系统中的运动目标回波信号模型构建,然后采用分数阶傅里叶变换进行运动目标信号分解和自适应匹配投影,引入四阶累积量切片作为后置处理器,在每一步搜索中进行自适应匹配投影滤波,利用四阶累积量解除时频耦合,抑制噪声,实现对运动目标信号的后置频谱聚焦,提高运动目标的检测性能。仿真结果表明,采用该算法进行智能激光夜视监控系统的运动目标检测,检测精度较高,准确检测概率加高,噪声抑制性能较好,性能优越。
In the intelligent laser night vision monitoring system for the detection of moving targets, the color noise received by the larger interference, resulting in the detection of moving targets decreased performance. The traditional method uses the fractional Fourier transform detection method, which will generate large false alarms in the detection of non-stationary moving targets for laser night vision monitoring, and the detection probability is not high. An improved fractional Fourier transform detection algorithm based on fourth-order cumulant slicing postprocessing is proposed to detect moving objects in the intelligent laser night vision monitoring system. The model of moving target echo signal in the intelligent laser night-vision monitoring system is constructed. Then the fractional Fourier transform is used to decompose the moving target signal and adaptively match the projection. The fourth-order cumulant slice is introduced as the post-processor. In the one-step search, the adaptive matching projection filter is used, the fourth-order cumulant is used to release the time-frequency coupling, and the noise is suppressed. The post-spectrum focusing of the moving target signal is achieved, and the detection performance of the moving target is improved. The simulation results show that the proposed algorithm can detect the moving targets of the intelligent laser night-vision monitoring system with high detection accuracy, higher detection probability, better noise suppression performance and superior performance.