Research On The Design And Application Of Dimple Depth And Eveness Model Based On Wavelet Transform

Through in-depth analysis of the steel material fracture failure mode, it has been found that fracture dimple is the fundamental basis for the morphology of fracture properties and also is the core basis of identification of fracture mechanism of materials. The depth and distribution eveness of the dimple have important influence on the properties of the material such as strong/hardness, plasticity and toughness,forming property and so on. The fracture mechanism, internal / external crack formation and the intrinsic potential of the main properties of the steel can be explained by the quantitative analysis of the depth and eveness of the fracture surface.Therefore, it can be concluded that the quantitative analysis of dimple depth and eveness is the crucial component of materials science and engineering field, and it has important significance of academic research and engineering practice value, but the dimple depth and its eveness measurement and classification technology is not yet reported at home and abroad.Main research contents and innovations are as follows:1.Analyzed specific steel dimple image, and combined with the existing image preprocessing algorithm, this paper presents a new wavelet threshold filtering algorithm based on wavelet transformation to remove dimple image noise and retain dimple edge information. Using image enhancement techniques to enhance identification of the dimple images.2.In view of the existing target image edge detection method of dimple characteristic has image edge deletions and fuzzy image defect which can not be detected, this paper use wavelet-fusion-algorithm of the wavelet transformation modulus maxima analysis method and improved Canny edge detection method. The dimple edge can be clearly displayed and detected by the fusion method which is based on the wavelet analysis and the morphological characteristics. A new pseudo target elimination algorithm is proposed to eliminate the false target phenomenoncaused by the precipitation of particles in the dimple image.3.To resolve the problem in dimple depth measurement, this paper extract a new concept of equivalent dimple depth from a large number of dimples images and build a calculating formula. The results demonstrate that 95% of equivalent depth is similar to the real dimple depth which show that the equivalent dimple depth can completely replace the real dimple depth for accurate analysis.4. According to the problem in the distribution of steel dimple depth to quantitatively characterize, this paper proposes a new depth eveness of dimple image based on feature parameters and measurement and classification method. This method which achieved precise characterization and scientific classification can achieve the eveness of steel dimple depth and can control the error under ? 0.001 um to obtain the best degree of the current steel dimple quantitative analysis.Based on the above research, through the application of a large number of engineering examples, the results show that this method can realize accurate measure and classify the dimple depth. The measurement and classification results can not only help technical personnel improve the existing preparation process, but also provide a reliable basis for the development of new steel.