Research On Iterative Reconstruction Algorithm Of Wood Stress Wave Nondestructive Testing Image

With the development of science technology and the implementation of China’s natural protection forest projects,wood non-destructive testing technology can detect the internal defects of wood without damaging the wood,which has a crucial role in protecting forest resources and improving the utilization of wood.Because the stress wave nondestructive testing technology is not harmful to the inspector,and has the advantages of a portable bag,high detection accuracy and suitable for field work,the stress wave based wood nondestructive testing technology has research value.Researches on two-dimensional image reconstruction of stress wave wood defects are divided into two directions:analytical algorithm and iterative algorithm image reconstructionmainly studies the reconstruction of two-dimensional wood defect images based on iterative algorithms.Aiming at the ill-posed problems in wood defect image reconstruction technology,this paper studies iterative algorithms and improves the speed of iterative convergence to make it easier to observe the position and shape of wood defect images.Studies the image reconstruction of wood defects.First,according to the wave theory of stress waves and the propagation law of different media in the wood.Stress waves are reflected and transmitted in different media.This paper introduces the ray tracing principle of bending method and test firing method,analyzes the advantages and disadvantages of different ray tracing methods,and finally selects the shortest path ray tracing method to determine the minimum travel time of stress waves in the wood.Based on the basic principle of ray tracing,an iterative algorithm numerical model of time,distance and velocity of stress waves is established.The reconstruction of wood defect images based on stress waves is an ill-posed problem,and iterative algorithms have a good effect in dealing with ill-posed problems.This paper studies and analyzes Tikhonov regularization algorithm,back projection method,ART iterative algorithm,and Landweber iterative algorithm.It is concluded that Landweber iterative algorithm has high stability in solving ill-posed problems,but the algorithm has low convergence speed.On this basis,A regularization iteration gain factor was proposed to improve the convergence speed of the Landweber iterative algorithm.In this image reconstruction experiment,12 stress wave sensors are uniformly embedded in the samples of Quercus mongolica and Pinus koraiensis,and the stress wave detection instrument is used to obtain the propagation time and distance of the stress wave in the log under normal temperature conditions,and is substituted into the improved Landweber iterative algorithm.The numerical model is used to calculate the slowness distribution matrix using MATLAB programming and realize the reconstruction of wood defect images with Sufer8.0 drawing software;discuss and analyze the impact of different tree species and different defect positions on the accuracy of reconstructed images,the convergence speed of different algorithms,and reconstructed images The analysis of the fit degree proves the feasibility of the algorithm.Finally,image threshold segmentation,image morphology processing and image edge detection make the reconstructed wood defect image more intuitive and accurate.The improved Landweber iterative algorithm is more accurate than the ART iterative algorithm in reconstructing images.Therefore,the iterative algorithm in this paper is suitable for wood defect image reconstruction and has practical significance.