The Research Of Detection Method About Grain Inner Holes’ Distribution Based On Multi-view Projection

In the past X-ray nondestructive testing of the grain inner hole, CT imaging systemobtain the projection data by scanning grain under different angle projection and using certainreconstruction algorithm to reconstruct the required information of grain inner hole. Inpractice, grain cannot be scanned under the complete angle due to the restriction of thescanning device and the object shape so that most of the projection data is incomplete. Thereconstruction algorithm about incomplete projection data also has certain limitation becauseof projection data completeness. This article designed the detection method and system aboutgrain inner holes distribution based on multi-view projection according to the microfocusX-ray imaging system due to the characteristic which the grain is small in size and inner holeis well-distributed.Firstly, the paper proposed the detection system about grain inner holes distributionbased on multi-view projection by analyzing the principle of X-ray imaging and CT imagingsystem and analyzed the imaging quality of microfocus X-ray source and the selection oflinear array detector. Secondly, the paper introduced FBP reconstruction algorithm, ART andTV-ART iterative reconstruction algorithm in detail based on the theoretical basis for CTreconstruction algorithm and reconstruct simulation model of the grain. The reconstructionresults were compared under the condition of different sampling interval and magnificationratio of the ART and TV-ART reconstruction algorithm so that the limitation of reconstructionresults is inferred under the less projection data. Finally, based on the above limitations ofreconstruction algorithm, the paper put forward the detection method about grain inner holes distribution based on multi-view projection. The algorithm process was described in detailthrough simulation experiment and the actual projection data of grain inner hole was detectedby using it. The eccentric problem of grain inner hole is made further analysis so as to verifythe feasibility and accuracy of the algorithm and confirm the rationality of the detection system.