Font Size: a A A

Research On Feature Extraction Methods For Ionization Tracks In Track Detectors

Posted on:2018-10-04Degree:DoctorType:Dissertation
Country:ChinaCandidate:T L LiFull Text:PDF
GTID:1362330566987983Subject:Nuclear Science and Technology
Abstract/Summary:PDF Full Text Request
With the development of high-density readout electronics and time projection chamber detectors,it is possible to measure complex ionization tracks(for example,lowenergy electron scattering tracks)using a single track detector.It is one of the hotspots in the current physical experiment to use the track detector to obtain the morphological and energy characteristics of the ionization track and to develop the corresponding feature extraction method.However,it is difficult to extract the track features due to the largeangle deflection and rewinding that may occur in the ionization track.In this paper,a general method for the feature extraction of ionization tracks is studied.With the knowledge of graph theory,a general feature quantity reflecting the morphological and energy characteristics of the ionization track is proposed,and successfully applied in the data analysis of the track detectors in different types of physical experiments.In this paper,the measurement process of the track detector is analyzed,and the general physical model of the track detector is established.The loss of the track features in the measurement process is discussed in detail,and the main objectives of the track feature extraction are summarized: reconstruction of trend features and reconstruction of energy features.In this paper,a new track feature extraction algorithm and the corresponding feature quantity "REF Track(Refined Energy-Filtered Track)" are proposed based on the physical model of the track and the shortest path problem in graph theory.The simulation and experiment results show that the REF Track can well restore the morphological and energy characteristics of the original track,and can be used for track reconstruction and event screening in physical experiments to optimize the performance index of these experiments.In this paper,the proposed track feature extraction algorithm and the feature quantity "REF Track" are applied to the 0??? experiment and astronomical X-ray polarimetry experiment respectively.In the 0??? experiment,the REF Track is used to extract energy characteristics,and a new event screening method is proposed from the energy deposition models of 0??? events and background events.The method can be applied to the 0??? experiments which are capable to measure particle tracks,and will contribute to the event identification and background reduction.In the astronomical X-ray polarimetry experiment,a new method based on the reconstruction of the REF Track is proposed,which can locate the interaction point the photoelectron more accurately and precisely,and better reconstruct the emission angle of complex photoelectron tracks.Justified with simulations and measurements,this new method significantly improved the modulation factor of the X-ray polarimetry in the high energy region,and does not introduce detectable residual modulation or systematics.
Keywords/Search Tags:0???, X-ray polarimetry, track detector, track reconstruction, feature extraction
PDF Full Text Request
Related items