张占文
摘 要: 当前的地物识别方法对地物反馈波特征掌握不准确,识别效果不好,对此,提出基于多普勒天气雷达的地物识别方法。将天气雷达信息点分为垂直形态、水平形态、纹理形态、垂直速度、水平速度、谱宽速度六个特征,通过查阅历史资料并结合反馈波的多普勒效应曲线,统计地物与水域特征差别。对地物与水域的折射率进行修正,根据目标物体与地物反馈波频率的不同,將混合在地物识别结果中的水域和目标物体提取出来,增加地物识别准确率。实验结果表明,所提方法拥有很好的识别效果。
关键词: 多普勒天气雷达; 多普勒效应曲线; 地物识别; 水域特征
中图分类号: TN95?34; P412.25 文献标识码: A 文章编号: 1004?373X(2017)19?0112?03
Research on ground?object recognition method based on Doppler weather radar
ZHANG Zhanwen
(Luoyang College, Civil Aviation Flight University of China, Luoyang 471000, China)
Abstract: The current ground?object recognition method has inaccurate mastery for ground?object feedback wave feature and poor identification effect, so a ground?object recognition method based on Doppler weather radar is put forward. The weather radar information points are divided into the vertical form, horizontal form, texture form, vertical speed, horizontal speed and spectrum width speed. The historical data is looked up and the Doppler effect curve of feedback wave is combined to survey the 7feature difference between ground?object and water area, and correct the refractive indexes of the ground?object and water area. According to the feedback wave frequencies difference between the target object and ground?object, the water area and target object mixed in the ground?object recognition results are extracted to improve the recognition accuracy of ground?object. The experimental results show that the proposed method has better recognition effect.
Keywords: Doppler weather radar; Doppler effect curve; ground?object identification; water area feature
0 引 言
天气雷达是勘察气象信息以及地物信息的一种现代化技术手段,由于气象信息的反馈波会严重影响雷达勘察精度,并且地物信息常常与目标信息混合在一起,对信息定量分析带来了不小的干扰,因此,在天气雷达输出目标信息之前,必须对气象信息和地物信息进行有效识别[1]。近年来,一些学者开始研究智能化的气象信息和地物信息识别方法,提出的空间反馈波检测理念和梯度检测理念都能够很好地识别出气象信息,但对地物信息的有效识别仍未取得较大进展。
最早的地物信息识别方法采用梯度检测反馈波,通过反馈波的伸缩和反射识别地物,忽略了地物反馈波速度以及水平形态的变化。文献[2]提出的基……