Research On Image Processing Algorithm In Testing Mechanical Properties Of High Temperature Materials

The mechanical properties of high temperature materials mainly include thermal strain,thermal expansion coefficient,elastic modulus and Poisson's ratio.The traditional contact measurements can not detect the properties of materials in the environment where the temperature exceeds 1000 °C,but the optical measurement method has the advantages of high accuracy and non-contact,and can be applied to high temperature environment detection.This paper aims to study the properties of material mechanics under high temperature conditions,through high temperature edge detection algorithm and digital image correlation algorithm to detect the thermal expansion coefficient and tensile strain of the tested specimens.This paper introduces the definition of image edge,and introduces the research of high temperature edge detection algorithm and digital image correlation algorithm at home and abroad.In the pixel-level edge detection algorithm,the process of Canny operator edge detection is introduced.In the sub-pixel edge detection algorithms,the Gaussian curve fitting algorithm,bicubic interpolation algorithm and gray moment algorithm are introduced and compared.Among them,gray moment algorithm has relatively high accuracy in edge detection.In order to detect the edge of the specimen in high temperature environment,this paper proposes an algorithm combining mathematical morphology edge detection algorithm and improved gray moments.Firstly,Canny operator is used to roughly locate the edge of the image.Then,a single effective pixel-level edge is obtained by using mathematical morphology edge detection algorithm and quadratic Canny operator.Finally,an improved gray moment algorithm is used to precisely locate the edge on the basis of the pixel-level edge.The experimental results show that the accuracy of the proposed algorithm can reach 0.1-0.15 pixels.This algorithm can realize high precision non-contact measurement of specimen edge under high temperature.Fast normalization algorithm and inverse combination Gauss-Newton algorithm are used in high temperature digital image correlation algorithm.For each corresponding point in the deformed image,the initial value is substituted into the inverse combined Gauss-Newton algorithm to speed up the single point matching rate,and the buffer interpolation coefficient is adopted to speed up the matching rate of the whole field deformation.The improved algorithm is compared with the surface fitting algorithm,the forward additive Gauss-Newton algorithm and the inverse combinedGauss-Newton algorithm in terms of precision and running time.The results show that the algorithm in this paper works best.This paper describes the difficulties in edge detection of high temperature image,and gives the design scheme of camera imaging and system,and sets up a high temperature test system in the laboratory.The high temperature edge detection algorithm and the improved inverse combination algorithm proposed in this paper are used to test the feasibility and accuracy of the two algorithms by thermal strain experiment and high temperature tension experiment.