The Design And Research Of Precipitates Measurement And Shape Classification System In Steel

In the research and production filed, it is often need to analysis the movement rule of precipitates in materials in different technical. As the great development of Ultra-fine grained steel, the size, shape and distribution of precipitates in steel have mostly influenced the capability of steel. In the same time, the precipitates in steel has change to nanoparticles from big inclusion. The measurement of precipitates’ s size ,shape and distribution in steel has become hot spot in the Ultra-fine grained study field. It not only involves many frontier fields of scientific research, such as material preparation、image processing、pattern recognition and etc, but also construct high-tech platform which has challenging more.The topic of this thesis is about the micro structure of the nanoparticles in steel. For the reason of the problem such as precipitate’s conglomeration, precipitates’s holes and precipitates’s burrs ,and the traditional manual statistic work’s low veracity and efficiency, Automatic measure and classification method based on morphological features and neural networks is proposed. The method has colligate many techniques such as image process, morphological theory and neural network theory.The method firstly get the image of precipitates in steel’s micro structure through the electron microscope, then smooth the picture and modify the gray scale of picture to increase it’s contrast and remove noise in the background. On the contrast of traditional image segment arithmetic, a multi-threshold based on area partition segment arithmetic is proposed. Aim at the problem of precipitate’s conglomeration, precipitates’s holes and precipitates’s burrs which are caused by irregular manual manipulation or precipitates’s vice , the method has used various morphological arithmetic and improved seed-fill arithmetic to solve this problem. According to a great deal of statistic and analysis work, the method has extracted the six Characteristic such as area, girth, the ratio of width and height, roundness rate and etc, to represent the shape of precipitates. Then mapping the relationship between morphological features vector and shape of precipitate through BP neural network to achieve the purpose of automatic classification statistic of precipitates.The experiment results show that, automatic measurement and classification method is excellent to solve the problem such as precipitate’s conglomeration, precipitates’s holes and precipitates’s burrs .Otherwise, the method can work efficiently and conveniently, it provides compellent evidence to the quantity analysis of precipitates in steel, and also can used in many other scientific fields.