太阳射电Ⅱ型爆发中频率分裂现象的特征自动提取研究

Research on Automatic Feature Extraction of Frequency Splitting Phenomenon in Solar Radio Type Ⅱ Burst

在射电动态频谱图上,太阳射电爆发中的Ⅱ型暴通常具有缓慢漂移的基频和二次谐频辐射带,这些辐射带又具有带分裂结构。Ⅱ型射电暴的频率分裂现象是太阳射电频谱图像研究的重要现象之一。本文首先采用通道归一化方法和大津法对频谱图像进行预处理,然后通过形态学开闭重建定位出爆发区域。确定爆发区域后,采用高斯卷积平滑爆发,并运用边缘指示函数实现太阳射电暴的边缘检测,最后基于距离正则化水平集方法提取出频率分裂现象。成功提取出频率分裂现象后简要介绍了特征参数的统计与分析方法,以便为进一步理解Ⅱ型射电爆发的机制提供依据。

On the radio frequency spectrum,the type Ⅱ storm in the solar radio pulse train usually has slowly drifting fundamental wave and second harmonic radiation bands,and these radiation bands have a band-split structure.The frequency splitting phenomenon of type Ⅱ radio burst is one of the important phenomena in the research on solar radio spectrum images.This article first uses the channel normalization method and the Otsu method to preprocess the spectrum image,and then locates the burst area through the morphological opening and closing reconstruction method.After locating the burst area,Gaussian convolution is used to smooth the burst,and the edge indicator function is used to realize the edge detection of the burst.Finally,the frequency splitting phenomenon is extracted based on the distance regularization level set method.After successfully extracting the frequency splitting phenomenon,the statistical and analysis methods of characteristic parameters are briefly introduced to provide a basis for further understanding of the mechanism of type Ⅱ radio bursts.