论文部分内容阅读
复杂背景下低信噪比弱小目标的自动检测是当今目标自动探测研究尚未解决的一个难题。将一维均值反差作为一种不相似性度量应用于小目标检测的前期滤波处理中,有效地增强小目标信息、抑制了复杂的背景和噪声,并结合背景预测原理,实现了对小目标的快速检测。仿真实验表明:该滤波算法大幅提高了小目标图像的信噪比,保证了利用背景预测原理检测小目标的准确性;与基于二维均值反差的滤波方法相比,该方法对小目标形状的适应性更强,速度更快。
Automatic Detection of Weak Small Signal with Low S / N Ratio under Complicated Background is an unsolved problem in current automatic detection of targets. The one-dimensional mean contrast as a measure of dissimilarity is applied to pre-filter processing of small target detection to effectively enhance the small target information and restrain the complex background and noise. Combined with the background prediction theory, Quick check. Simulation results show that the proposed algorithm can significantly improve the signal-to-noise ratio of small target images and ensure the accuracy of detecting small targets by background prediction. Compared with the filtering method based on two-dimensional mean-contrast, Adaptable, faster.