Study On The Method Of Removing Stains From Pr/Nd Mixed Solution Image

Rare Earth is known as "industrial vitamins" it is the most important strategic resource in today’s society. China is one of the few countries which contain rich rare earth resources, the rare earth separation production scale and product quality has an important position in the world. However, our country’s rare earth extraction separation process is still in the stage of off-line analysis, empirical adjustment and manual control, which leads to a lot of waste of resources and the quality of products is not stable and so on. The rare earth element component content online detection is a key step of the automatic control of China’s rare earth extraction process.In the rare earth extraction process, because of some rare earth ions with ion color special features, making use of machine vision technology of rare earth element component content detection becomes possible. According to the special color of praseodymium and neodymium ions ion characteristics, The laboratory has developed a rapid detection system based on machine vision for the content of rare earth elements, Is solved with special ions characteristic color of praseodymium and neodymium extraction separation process elements group content rapid measuring in an effective way. Provide technical suppor for the Pr/Nd extraction process optimization operation. It has extremely important theoretical and practical significance to realize the sustainable development of China’s rare earth extraction separation industry. In paper, the main research work and innovations are as follows:(1) Based on the two aspects of hardware system design and software system design analyzed the rapid detection system of rare earth element component content based on machine vision. The hardware design of the system, Analysis of the factors of the basic principle of selection of camera, lens, image acquisition card, light source, computer hardware and the necessary attention. The software system design, Introduces the method of mixed programming between MATLAB and LabVIEW, Analysis of the rare earth element content rapid detection system of the whole structure and the basic working principle of image acquisition module, image processing module, the display module,the other modules in system.(2) According to the laboratory of rare earth element components can only be detected in the laboratory environment, to detect the content of the rare earth solution image without stains, it can’t adapt to the field of rare earth extraction process. Based on FCM algorithm for image segmentation algorithm of detection system of image processing part of corresponding improvement and make the detection system capable of detecting with stain Pr/Nd solution image corresponding to the component content value. Analysis of the derivation of FCM algorithm in clustering criterion, using the Lagrange multiplier method for solving membership matrix and clustering center function. Introduces the method of determining the number of cluster categories c and the method of fuzzy weighted index m and the implementation steps of FCM image segmentation algorithm. By using the method of Euclidean distance and spatial distance combination calculation between pixel and cluster center distance.(3) In Jiangxi a rare earth extraction enterprise in the Pr/Nd solution as the research object. Analysis of the conversion between RGB and HIS color space method, determine the containing stain praseodymium neodymium solution image color feature extraction method.The use of HSI color space of the H, S and I feature of first order moment description of the rare earth mixed solution of image color information. FCM algorithm to process the collected Pr/Nd mixed solution containing the image stains, get the two value image gray target solution. Using the gray value of two images to find the target solution area of the original color image in the pixel, and calculates the color spatial feature value and component content value. The content of the component detected by the Pr/Nd solution image without stain points is a reference. Results show, The FCM clustering algorithm can effectively inhibit the stain of color spatial feature value extraction and content determination of, for rare earth elements group is divided into content detection system can better application in the field of rare earth extraction process has laid a theoretical foundation.