Research On Quality Inspection Technology Of Diamond Production For Abrasives

With the large-scale application of diamond,synthetic diamond has developed rapidly,and it has become a reality to synthesize diamond in large quantities in industry.Henan is a big province of synthetic diamond enterprises.90% of synthetic diamonds in the world are made in China,80% of the 90% of the production is produced by Henan enterprises.,which has the saying that " synthetic diamonds look at China,and Chinese diamonds look at Henan." The quality of synthetic diamond is affected by the synthesis temperature,and the synthesis condition of diamond are high temperature and high pressure.Under such harsh conditions,it is impossible to directly measure the cavity temperature of synthetic diamond,and only the quality of synthetic carbon diamond particles can indirectly reflect the suitability of the synthesis process.Because diamond synthesis companies use traditional methods such as manually relying on microscopes to observe the quality of synthetic diamonds empirically.This detection method is susceptible to subjective,empirical,visual fatigue and other factors,leading to an increase in false positive rate.The paper applies machine vision technology to the quality inspection of abrasive particles of abrasives.The percentage of diamond content in carbon rod and the characteristic parameters of synthetic carbon rod such as diamond particle size,crystal form,clarity,roundness and ellipticity are important indicators for feedback the suitability of the production process and the accurate measurement is the focus of this paper.In this paper,the traditional image binarization method is introduced.Aiming at the problem of poor preprocessing and even failure of the diamond particle image in the carbon rod by the traditional binarization method,the image segmentation threshold is dynamically selected based on the rank-and-column two-mean interlacing method,so that the purpose of binarizing the diamond particle image in the carbon rod is achieved,and the determination of diamond content in the synthetic carbon rod is realized.In the aspect of determining the crystal form of diamond particles,it is difficult to determine the crystal form for the traditional image processing method.The deep learning technology is proposed to be applied to the detection of diamond particle crystal form of synthetic carbon rod section;the software developed by machine vision technology can effectively detect the quality index of multiple diamond particles,and can save the measurement data in real time,which is convenient for data accumulation,and provides data support for reflecting process,evaluating process and improving process.