论文部分内容阅读
针对红外序列图像中运动弱小点目标的检测问题,设计了一种基于改进遗传算法优化的修正Top-Hat形态学滤波器算子。其中,优化的修正Top-Hat形态学滤波器可以很好地抑制背景和噪声的影响;改进遗传算法采用新的区间离散化编码和自适应的主次式交叉与变异算子,通过优化搜索全局空间得到的形态学滤波器参量具有较好的滤波性及时效性。并且针对不同信噪比的点目标检测建立了自适应门限。实测数据的处理结果表明:在虚警概率小于5%情况下,优化的修正Top-Hat形态学滤波器算子对信噪比约为2的复杂图像检测概率大于等于70%,与固定结构元素的Top-Hat形态学滤波器相比检测概率提高了近10%,与用经典遗传算法训练的传统Top-Hat形态学滤波器相比检测概率提高了4%。
Aiming at the detection of moving weak dot targets in infrared sequence images, a modified Top-Hat morphological filter operator based on improved genetic algorithm is designed. Among them, the optimized modified Top-Hat morphological filter can restrain the influence of background and noise well. The improved genetic algorithm uses the new interval discretization coding and the adaptive principal and sub-intersection and mutation operator, The morphological filter parameters obtained in space have better filtering and timeliness. And for different signal to noise ratio point target detection established adaptive threshold. The processing results of the measured data show that the optimal modified Top-Hat morphological filter operator has a detection probability of 70% or more for the complex image with signal-to-noise ratio of about 2 when the false alarm probability is less than 5% Top-Hat morphological filter increased the detection probability by nearly 10% compared with the conventional Top-Hat morphological filter trained by classical genetic algorithm, and the detection probability increased 4%.