| 其他摘要 | The sun is the closest star to the earth and provides the light and heat necessary for life on the earth. In the meantime, the sun is also the unique star which can be observed and studied nearby. Currently, a variety of instruments and methods have been used in order to study the sun. Nevertheless, in the studies, magnetic field plays an essential role in that the disaster weather in interplanetary space and the formation and occurrence of various characteristic structures and activity phenomena of photosphere, chromosphere and the of corona of the sun are closely related to the solar magnetic field. Therefore, it is of great significance to diagnose the solar magnetic field from the photosphere to the chromosphere and then to the corona. But, due to that the corona is optically thin and its radiation intensity is weak very much, to measure its magnetic field reasonably, the conventional method is to extrapolate it by taking the photosphere magnetic field obtained by observation and inversion as the boundary condition. The extrapolation methods includes potential field method and force-free magnetic field method. Unfortunately, many shortcomings exist in the conventional extrapolation methods, resulting in the enormous differences in configurations between the extrapolated magnetic field and coronal loops which are the intuitive representative of the magnetic field of the corona. Currently, some researchers have proposed optimized methods so as to decrease the differences and obtain more reliable magnetic field of the corona, in which detected coronal loops often are used as the constraint conditions. Therefore, the efficient detection of corona loops in the coronal images is a prerequisite for obtaining more reliable magnetic field of the corona. On the other hand, in various research subjects associated with the sun, the problem of coronal heating has long puzzled solar physicists, and to solve the problem, solar physicists have proposed various coronal heating mechanisms and models and have adopted various methods to verify which mechanisms and models are effective. However, so far, there is no agreement on which mechanism and model are more reasonable. It is worth noting that in mechanisms and models, coronal loops play an important role. In particular, for coronal loops, the study of their parameters, such as length, width of cross-section profiles, lifetime, density and temperature, is significant for the understanding of the mechanisms and models of the coronal heating. Fiber Arrayed Solar Optical Telescope, FASOT, is a novel and dedicated solar telescope with high scientific and observational goals designed as specially optical and mechanical structures. Its main scientific goal is the real-time and high accurate measurement of the magnetic fields of solar photosphere and chromosphere (a priority is given to the magnetic field of the chromosphere). Furthermore, due to the special optical and mechanical structures of FASOT, to implement its scientific and observational goals, the dedicated observational control system is needed. On the other hand, the future scientific data of FASOT will be the high-precision magnetic fields of photosphere and chromosphere of the sun. However, to extrapolate the more reliable magnetic field of the corona based on the scientific data of FASOT and implement the diagnosis of the solar magnetic field from the photosphere to the chromosphere and then to the corona, it is the prerequisite to efficiently detect coronal loops contained in corona images. But, in the corona images, there are many coronal loops, and the boundaries between them are ambiguous, resulting in that it is difficult to detect them using traditional artificial vision detection and image processing methods. Therefore, it is significant to develop an efficient automated detection algorithm of coronal loops, which can not only provide application supports to the future scientific data of FASOT, but also help us to study the problems of the coronal heating associated with coronal loops by a large number of sample. Based on the descriptions mentioned above, following researches will be carried out:(1) A dedicated observational control system is designed so as to meet the specially optical and mechanical structures of FASOT and verified in the second generation prototype of FASOT, FASOT-1B. The observational control system is an integrated system which has the functions of auto guiding, the scanning of two-dimensional (2D) integral field of view, the rotating of the gratings of spectrometers, the switching of polarization correction wheels and the controlling of polarization modulation and data acquisition. With the help of the observational control system, FASOT can finish its polarization observation and the expansion of the field of view. In particularly, for the scanning function of two-dimensional integral field of view integrated into the system, this paper first puts forward using the closed loop feedback control of auto guiding to guide the mount of FASOT, driving the simultaneous movement of its guiding optics and post-focus instrument. In the meantime, in the scanning process, to enhance the stability of scanning, the Kalman filter is used to suppress the random disturbances forced on the control quantity of closed-loop feedback (the coordinates of the center of gravity of the solar images). (2) An automated detection and extraction algorithm of coronal loops based on match filter and approximate local directions and connected domains of pixels is proposed and its performance is compared with present algorithms, especially the two algorithms with the most outstanding performance, (Oriented Coronal CUrved Loop Tracing, OCCULT) and its improved version, OCCULT-2, in order to show that the proposed algorithm can indeed work satisfactorily and provide technology supports to the future scientific data of FASOT.(3) Based on the proposed algorithm of detecting and extracting coronal loops, the coronal loops contained in images of 193 Å waveband of (Atmospheric Imaging Assembly/Solar Dynamics Observatory, AIA/SDO) and High-resolution Coronal Imager (Hi-C) are automatically detected. In the meantime, the widths of the detected coronal loops are automatically calculated and statistically analyzed. The statistical results demonstrate that due to the lack of the statistical sample of the fine-structured coronal loops, the statistical results of the widths of coronal loops based on the algorithm, OCCULT are incomplete and in comparison with the high resolution of Hi-C, the resolution of AIA/SDO cannot distinguish most of fine coronal loops, which can help solar physicists better model corona heating. |
修改评论