摘要
介绍了多频谱红外烟幕干扰弹的反制条件和反制时机;在可反制的阈值范围内,对烟幕遮蔽下的目标进行图像增强等预处理后,采用小波多分辨分析和分形布朗随机场模型,计算分形维数和分形拟合误差,从而判断目标所在区域;最后采用最佳熵自动阈值法确定分割门限,将目标从烟幕遮蔽中分割出来。实验证明,该方法效果明显,是对多频谱红外烟幕干扰弹反制最为关键的一步。
This paper introduces the condition of countermeasures of the multispectral infrared interference bullet. It uses the image enhancement in the threshold of the resisted scope. In order to judge the region of the objective place, this paper uses the wavelet transform and fractal theory to compute the fractal dimensions and the fitting errors. Then it uses best entropy that computes the numerical value of threshold to divide the target from the infrared image. The experiment proves that the means of this paper have good effects, and it is the critical step for the countermeasures of the multispeetral infrared interference bullet.
出处
《激光与红外》
CAS
CSCD
北大核心
2008年第11期1110-1113,共4页
Laser & Infrared
关键词
红外烟幕干扰弹
图像增强
小波分析
分形布朗随机场
图像阈值分割
infrared smoke interference bullet
image enhancement
wavelet analysis
fractal model
threshold value divided
