Researches On The Design And Its Application Of Dimple Morph Analyze Model Based On Wavelet Transform

In a variety fracture failure analysis of iron and steel materials,the fracture mechanism can be drawn through micro-analysis on the fracture structures(dimples),and then the impact essentials of materials on mechanical properties can be proved.Therefore,precise analysis on dimples, especially accurate and efficient analysis on ultra-fine materials dimples has important theoretical study significance and engineering application values. However,for a long time,the micro-analysis on dimples of steel rely on the professionals and manual statistics methods,which inevitably led to low efficiency,low accuracy,as well as use up a lot of human resources,and seriously affected the following material quality control work.The present work was funded by the National Nature Science Foundation of China.The dimples in steel were used as research targets and established the steel dimple measuring and classification system for the first time based on wavelet analysis.The system mainly uses the B-spline wavelet transform and morphological approach to segmentation and processes the target images,then realized the automatic measurement and classification of dimples.Main contents and innovations:1.First a comprehensive analysis and studies has carried on a variety of image processing algorithms based on analyze of steel dimple image characteristics.And carries on the image pretreatment using the wavelet transformation,eliminates each kind of image noise.2.In order to overcome the edge missing and ambiguous problems caused by the traditional edge detection methods,an improved wavelet edge detection method based on B-spline wavelet was proposed.The new method has made a fully integration of wavelet analysis and morphological methods, can test and show the edge of steel dimples completely and clearly.3.An adaptive method was introduced to threshold based on index of fuzziness to divide dimple from background,and eliminated the fuzzy characteristic which in the nest image existed,simultaneously retained the dimple edge feature.4.According to image characteristics of steel dimples,the recursive marking algorithm and morphological formulas was selected by this paper and used to calibrate and measure the dimples,and then generates a corresponding analysis and measurement report.Finally,by carries on a large number of analysis and tests on the system,the experiment results indicated that the steel dimple measuring and classification system based on wavelet analysis can test and show the edge of steel dimples completely and clearly.The system can carries on accurate, efficient and convenient measurement and classifications on dimples,thus provide reliable basis for analysis the relations between microstructures and the forming properties of materials.