| During the production process of the ceramic components,due to the limitations of the production environment and the production process,cracks or defects appear on the surfaces of some of the produced components,that is,unqualified products.Products with cracks and defects have a certain safety hazard in the process of fuel combustion.If components with cracks or defects on the surface can't be removed in time,there will be some safety hazards in the use of subsequent components.In order to eliminate potential safety hazards and remove the unqualified products,most of the samples are manually removed by visual inspection.However,artificial visual inspections have the disadvantages of high subjectivity,low efficiency,and high labor costs.Therefore,there is an urgent need for an efficient detection method to replace the completion of surface integrity testing of ceramic elements by humans.In recent years,digital image processing technology has developed rapidly.The automatic detection technology based on image processing has been widely used in many fields due to its advantages of non-contact,high precision,and high speed.In this paper,by using the relevant knowledge in the field of image processing,taking the ceramic elements as the objects,the method of detecting cracks and defects on the end face of the components based on image processing is studied.In this article,the main tasks studied are as follows:(1)An image acquisition system for crack detection of ceramic elements was built.The low-angle annular light is used as an illumination source to photograph the end-face image of the component,and the collected image is used as the original image for crack detection.(2)An image acquisition system for defect detection of ceramic elements was built based on dual-stripe structured light.The system uses stripe structure to light the left and right way,to complete coverage of defects on the end surface of the component.In the captured image,the structure of the defective part of the light stripe has a significant change in direction,for the extraction of the subsequent defect area provides a clear feature basis.(3)This paper present a new crack detection method for the finer surface crack and the non-uniform gray values of micro-ceramic components.The proposed method consists of three steps.First,a double threshold method based on bilateral filtering and local gray difference processing is used to acquire candidate cracks.However,because the gray value of the crack region is not continuous,a complete extraction may not be possible for the same crack,and only some crack fragments can be obtained;then,based on the proximity and continuity characteristics of candidate crack segments,tensor voting is used to obtain the end face.The significant pattern of each crack and the complete crack curve;finally through the crack curve combined with the previously extracted candidate crack fragments,the crack fragments belonging to the same crack are connected and the non-crack areas are removed.The experimental results show that the proposed algorithm has better performance than the method based on the connection of crack fragments.(4)For the feature that the structural light stripe changes in the defect part of the image,this paper proposes a method of using the self-fitting template line to mark the stripe deviation point in the image to obtain the surface defect.The method first uses the Steger algorithm to extract the center line of the structured light fringes;afterwards,according to the four conditions of the line segment connection,the concentric center line segment is connected and the connected center line is fitted by the least squares method.The template line is obtained;the stripe deviation point is marked according to the template line,and the deviation point of the mark is subjected to morphological processing to obtain the defect area. |
