论文部分内容阅读
随着激光技术的发展,激光成像雷达在现代战争复杂战场环境中逐渐获得了广泛的应用,目前激光成像雷达目标识别技术已成为国内外研究的热点。基于激光成像雷达的距离像,以地面军事目标为识别对象,提出了针对奇异值特征的激光成像雷达目标识别算法。利用奇异值特征表征图像具有良好的稳定性,通过分析识别率与奇异值特征数目的关系,从激光成像雷达目标的距离像中提取奇异值特征,该特征有效抑制了噪声,最后应用最优参数支持向量机对3种地面军事车辆目标进行识别。仿真实验表明,该算法取得了比较高的识别率,识别效果良好。
With the development of laser technology, laser imaging radar has been widely used in complex battlefield environment of modern warfare. At present, laser imaging radar target recognition technology has become a hot spot both at home and abroad. Based on the distance image of laser imaging radar and the target of military ground recognition, a target recognition algorithm for laser imaging radar based on singular value features is proposed. The singular value feature is used to characterize the image with good stability. By analyzing the relationship between the recognition rate and the number of singular value features, the singular value feature is extracted from the distance image of the laser imaging radar target. This feature effectively reduces the noise and finally applies the optimal parameter Support Vector Machines (SVMs) recognize three kinds of ground military vehicles. Simulation results show that the algorithm achieves high recognition rate and good recognition effect.