多普勒天气雷达下击暴流图像识别

以下击暴流具有的低层显著辐散特征为基础，引入图像识别中的连通区识别技术，开发出了应用于多普勒天气雷达的下击暴流图像识别算法。首先通过设置的速度阈值将径向速度进行二值化处理，然后运用8邻域法寻找速度大值区，并采用距离、夹角和正、负速度差值等条件进行约束对正、负速度大值区进行配对，最后对未成功配对的速度大值区进行邻近区域的二次匹配，从而识别出下击暴流区域。利用多个下击暴流个例实测的多普勒雷达数据对该算法进行了测试，结果表明该算法对一些较小尺度的下击暴流，尤其对受环境风场影响而具有不对称辐散特征的下击暴流有良好的识别效果。

Identification of Downbursts Based on WSR 88D Doppler Weather Radar Images

According to the significant low level divergence characteristics in the flow field of a downburst and the identification technology of the connected area in pattern recognition, an identification algorithm of downbursts using Doppler radar data is developed. The binary processing of radial velocity is carried out by setting the threshold of speed. The 8 neighborhood method is applied for searching big speed value zones, and the big positive and negative speed zones are matched by using constraints such as distance, angle, difference of positive and negative speeds, and so on. The secondary matching for the adjacent areas is conducted on big speed zones where matching failed, thus to identify the alert area of downbursts. Some real radar detection data are used to verify the algorithm. The results show that this algorithm can detect the small scale downburst with the clear low level divergence characteristic, especially for the downburst with asymmetric divergence characteristics that affected by the environment wind field.