Image Analysis On Cast Iron Microstructure Based On Digital Technology

All the time, most researchers take traditional approaches of qualitative or semi-quantity to evaluate the images of metal microstructure. However, artificial measurements may result in quite a number of errors. The efficiency and precise is low in this way. And its reliability, accuracy and reappeareance can not satisfy the requirement of quantitative analysis of metallographic microstructure.With the development of metallography, quantitative metallographic analysis technologies are to be employed in metallographic microstructure examining.Has selected cast iron as the researched materials.After crustification, rubbing, polishing, eroding procedures and so on, has prepared the test specimen. Observes the test specimen under the microscope, determines the best gathering field of view.Captures the images.It is necessary that applying the digital image processing technology to the cast iron metallography analysis to improve the precision and efficiency of detecting.Image analysis is the foundation of metallograph analysis.The image result affects the analytical speed and results precision.In order to improve the accuracy and efficiency of cast iron microstructure analysis, provides the study on microstructure image processing and analyzing methods based on digital image processing technology. Carries on the demarcation to the captured image.Transforms the image from the color space to the gradation space.Aiming at the non-uniform phenomenon of image brightness, proposes one method based on the theory of homeostasis filtering.Does experiments on commonly smoothing methods in both spatial domain and frequency domain.Uses the median filtering to remove the noise. Analyzes the histogram of cast iron gray image, using some segmentation methods to segment the cast iron microstructure image, such as threshold method,FCM method,Genetic method and Mathematic morphology method. And analyzes the segmentation results.Taking into account the merit of two dimension histogram, proposes to combine fuzzy cluster algorithm with genetic algorithm to segment the image based on two-dimensional histogram.Aiming at the limitation of single-scale algorithm of mathematics morphology gradient operator, proposes a multi-scale and multi-structure elements algorithm to realize the cast iron image segmentation. Compared with traditional edge detecting methods, the method can avoid the effect of noise and get much better edge results.Finally, picks up the area, perimeter, shape factor, ratio of the long and minor axis, extension and centroid of the labeled objects.Uses the area rate to compute the spheroidization rate, and carries out the rating of spheroidization rate of the modular cast iron. Integrates image analysis system, and save the results of feature parameters.