The sun is the nearest star to the earth, and the only star that can be accurately observed by human beings at present. The activities of the sun have a significant impact on human production and life. As a very active celestial body, the sun often has violent activities or bursts (such as flares and CME), which can be observed in both optical and radio bands. Solar radio bursts carry a large amount of important information related to solar physics. The high-energy particles and shock waves generated in the process of flare and CME spread in solar-terrestrial space, which may reach the earth and trigger a series of solar-terrestrial physical effects. It will have a very important impact on human aerospace engineering and communication facilities. Therefore, to effectively forecast and warn of space weather and reduce the damage of disastrous space weather to human production and life, it is necessary to monitor solar radio burst events in real time.In research on the automatic identification of solar type Ⅲ and Ⅱ radio bursts, this paper analyses and improves the existing detection methods of solar type Ⅲ and Ⅱ radio bursts, and proposes a new identification method. The main work is as follows:First, we preprocess the data of the solar radio dynamic spectrum, this includes dividing the data into several segments of length suitable for detecting solar radio bursts, and processing the data of each channel with uneven intensity; then, the solar radio dynamic spectrum is binarized and corroded, the purpose of binarization is to better identify the structure of radio bursts. During the process of binarization, attention should be given to the separation of foreground information and background information, eroding eliminates the random interference signals received in the process of receiving solar radio signals. Finally, the probabilistic Hough transform algorithm is used to find the linear structure conforming to the constraint of solar radio burst parameters in the binarized solar radio dynamic spectrogram data, and to extract the relevant information of the linear structure.This paper uses the solar radio observation data of the Nancy Decameter Array to identify solar type Ⅲ radio bursts and the solar radio observation data from the Meter-wave solar radio spectrometer of the Yunnan Observatory of the Chinese Academy of Sciences to identify solar type Ⅱ radio bursts. The experimental results show that the method proposed in this paper has high recognition efficiency and can more accurately extract information on the start and end times and frequencies of solar radio bursts. Its running speed also meets the real-time requirements of automatic detection of solar radio bursts.
修改评论