Study On Automatic Scaling Of E, Es In Ionogram Based On Image Processing

This paper introduces an automatic scaling algorithm of ionograms on thetechnology of image processing, and focuses on the preprocessing and the scalingalgorithm of E as well as Es. Having great influence on wireless communication,ionosphere is a part of atmosphere on earth, and it is demonstrated in papers specialdisturbance will happen to ionosphere before and after earthquake. Therefore itmakes much sense for human living to implement real-time monitoring ofionophere.Vertical sounding is the earliest means of sounding ionosphere, widely useduntil now. Nowadays there is more advanced ionosonde(the instrument for verticalsounding ionophere), so it is qualified to watch the ionosphere lively. Additionally,there are large quantities of accumulated sounding data to scale for the study onionosphere, because vertical sounding has so long a history. It has urgent needs forthe automatic scaling of vertical sounding data to reduce labour input.Vertical sounding data is usually transformed into ionogram, which contains theecho signal of ionoshpere. Considering ionogram is interpreted manually with itsimage characteristics, so this paper explores the automatic scaling of ionograms withimage processing technology. This paper discusses the image processing algorithmto recognize, classify and scale E and Es by geometrical and spatial position features.The work mainly contains:1.Systematically introduce the knowledge of ionosphere vertical sounding,including the definition of characteristic parameters of ionogram. Summarize andanalyze the existing automatic scaling means, laying foundation for the research onscaling algorithm. Listing the meaning of automatical scaling ionogram andmonitoring ionosphere, we confirm the research content and research sense of thissubject and propose the overall flow of algorithm. The proposed algorithm includes two steps:(1) The image preprocessing;(2) The recognition and scaling of E andEs.2.Analyze the existing scaling algorithm and point out the importance ofpreprocessing of ionograms. The image preprocessing method is proposed: Based onthe feature of gray-level and statistic virtual height of ionogram, divide trace regionin ionogram. Making use of the gray-level characteristic to enhance trace anddenoise in ionogram. Utilize the blank valley and analyse its position to divideionogram into two parts: E region and F region, scaled respectively.3.Analyse the shape characteristic of the trace in E region and classify them.Extract the characteristic position information of each trace class and transform theposition information into characteristic parameters with the scaling principal ofionogram interpretations. Additionally, some special ionograms and problems areanalyed.