Study On Image Reconstruction Algorithm And Image Processing System For CT Detection Of Explosive Hidden In Passenger's Baggage At The Airport

The explosive that conceals in passenger's baggage at the airport threatens the passenger's safety seriously, examining explosive accurately and quickly becomes a hot and challengeable problem. Due to huge advantage in the field of material detection, Computerized Tomography (CT) technology plays an important role in the realm of explosive detection. However, examination speed of the general CT machines is slow and the pass rate of baggage is slow too. Along with the increment of the passengers, CT security inspection equipments face a rigorous challenge. Therefore we only have to research the CT detection method and develop a sort of CT equipment which characterizes by high pass rate of baggage to solve the problem.In the process of CT detection, image reconstruction algorithm is the core problem. Firstly researching on detection of explosive at home and abroad is summarized,CT technology applied in the field of explosive detection is also summed; Then the filter back-projection(FBP) algorithm and algebraic iterative algorithm are discussed which are two of the major image reconstruction methods in CT, besides their performance are analyzed; The choice of window function which is the core of the FBP algorithm is studied, the thesis lucubrates the performance of various window functions; Aiming at the requirement of image quality in the CT detection of explosive, an improved window function is proposed; Combined the improved window function, the merit and demerit of FBP algorithm and iterative algorithm while look after both sides of quality and speed of the reconstruction image, we make use of a new method to reconstruct image; Aiming at noises in projection reconstruction system, to carry on analysis and make the summary and comparison toward de-noise methods, an efficient image-denoising method is proposed—combination with wavelet transform and fast median filtering, and a new threshold function is applied in decomposable coefficients, otherwise, the steps of the algorithm and the simulation results are also gave.