基于小波能量的自然纹理背景中多目标检测方法

Detecting of Targets in Natural Texture Background Based on Wavelet Energy

多目标跟踪技术被广泛用于军事领域。传统的目标分割方法,如基于数学形态学的方法和最大熵域值分割法,这些方法虽然能分割出目标,但往往与目标灰度相近的背景也被分割到目标图中,这对下一步的目标形心计算十分不利。根据自然背景和弱目标的频率特性,提出了一种基于小波能量检测目标的方法。目标在水平、垂直两个方向上的小波能量要远高于背景。再经过自适应域值选取,能够实现目标和背景完全分割。用分类法计算每个目标的形心。经仿真实验证明,该方法能够检测出海、空等自然背景中多个弱小目标。

The technology of targets tracking is widely used in military. Traditional targets segmentation methods, such as mathematics morphologic method, most entropy threshold method, which can divide targets from background, but the background which gray -level near targets is divided in the targets image. That is unfavourable for the figure center calculation of targets. According to the frequency characteristic of natural texture background and small targets, a method of detecting targets based on wavelet energy is suggested. The energy of targets in horizontal and vertical direction outclass that of background. Then through selecting adaptive region - value, targets and background can be divided completely. By the method of classifying, the figure centers of every target can be calculated. Experiment results shown that small targets in sea or sky nature texture background can be detected by this method.