Research On Quality Inspection System And Application Of Diamond Cutting Line Quality

Diamond cutting line is an important high-precision cutting consumable.It is obtained by attaching diamond particles on the metal substrate line.The substrate wire diameter is generally tens of microns,and the diamond particle size is generally several microns.Due to the small size and complicated production process,the quality control is generally performed by manual sampling.However,manual testing has the problems of high false detection rate,low detection accuracy,and cannot perform full inspection,and its quality control ability is average.This paper proposes a detection method based on machine vision to construct a diamond cutting line detection system to solve the problem of diamond cutting line detection.Diamond cutting line thickness,particle size,and particle distribution are more uniform and more consistent with its density requirements.The better the effect and quality of the cutting object,these are also important criteria for evaluating the quality of diamond cutting line.Aiming at the above detection problems,this paper intends to propose a machine vision-based diamond cutting line detection scheme.The main research contents of the paper are as follows:1.According to the special characteristics of the research object,design a reasonable lighting and imaging system,determine the parameters required for the camera,light source,lens,etc.,and compare the different lighting methods,select the appropriate lighting method,etc.,and select the appropriate lighting method.The imaging system collects images to ensure the acquisition of high-quality diamond cutting line images,selects a suitable software platform for the detection system,and introduces the software system and construction of VC ++ and Open CV configuration.2.Aiming at the problem that the surface of the diamond cutting line is easy to reflect light,the segmentation effect is reduced,and the accuracy of extraction is not enough,the method based on the combination of homomorphic filtering processing and edge filling is researched to eliminate the unevenness of light and disturb High-frequency signals areconverted into low-and medium-frequency signals,and the image is binarized using the maximum inter-class variance method.Based on the background estimation,the unevenness of the background is effectively eliminated,and the cylindrical uneven brightness surface image is realized adaptive segmentation.3.Aiming at the problem of particle adhesion in the research object,a self-labeled eight-connected domain method and an improved watershed segmentation method are proposed,which are fused with morphological processing,using morphological reconstruction,label calculation,watershed algorithm,and region merging,which are effective.The problem of over-segmentation in the traditional watershed method is suppressed,and the speed and accuracy of the eight-connected domain labeling are improved.After morphological opening and closing operations and adaptive label extraction,the segmentation effect is improved.The feasibility of the method was verified by experiments,and the accuracy of subsequent particle counting and particle basic parameter extraction was guaranteed.4.Based on the above,the parameters of the diamond cutting line are measured,including area,perimeter,density,etc.The particle size is measured according to the speed of movement and the exposure time,the circumference of the particles is calculated using the chain code method,the area is measured by the pixel counting method,and the density is measured by the kernel density estimation method.According to the parameters,the defects of the diamond cutting line particles can be obtained.Finally,the vector machine classification combined with the principal component analysis method is used to classify the defects according to the particle size and perimeter parameters.The distribution characteristics of the particles,agglomeration,excessive density,unstable sanding,Detection of sand and other conditions to achieve quality analysis.Through the research and solution of the above problems,a feasible and efficient detection path is provided for the detection of diamond cutting line defects,and it is expected that this detection technology will be applied to the detection process of diamond cutting lines to achieve online quality inspection of diamond cutting lines.