摘要
在分析红外图像弱小目标特征和背景特征的基础上,提出了基于小波变换和支持向量回归(SVR)的自适应滤波的检测方法.该方法首先采用小波变换抑制大部分背景杂波;然后用基于SVR的自适应滤波器(SVRBAF)对高频小波系数进行处理,大大提高了图像的信噪比;最后,基于目标的连续性和运动轨迹的一致性,采用流水线结构的序列处理方法进一步提高检测性能.仿真结果表明:该方法可显著提高红外目标的检测概率,实现较远距离弱小目标的检测.
By analyzing dim target and background characteristics from infrared images, a new detection method based on the wavelet transform and the support vector regression adaptive filter (WTSVRAF) was presented. With this method, the wavelet transform is used to inhibit the majority of background clutter. Then the SVR-based adaptive filter (SVRBAF) is adopted to treat high-frequency wavelet coefficients. As a result, the image signal-to-noise ratio (SNR)is greatly improved. Furthermore, based on the continuity of the target and the coherence of trajectory, the image sequence processing method is utilized to further improve the detection performance. The simulation results show that the detection method can significantly increase the probability of infrared target detection to achieve long-range target detection.
出处
《西南交通大学学报》
EI
CSCD
北大核心
2008年第5期555-560,共6页
Journal of Southwest Jiaotong University