Parameters Identification Of Grain Size Of Heap Leaching Uranium Ore Based On Image Analysis

Due to its rapid development, nuclear power now is in great demand of natural uranium. The development of uranium mining and metallurgy technology has gone into a new stage. At the present stage, the main means of extracting natural uranium is heap leaching technology. In order to save resources and maximize economic benefits, uranium mining and metallurgy enterprises must actively research heap leaching technology and improve its technological content. Ore granularity is the key factor which affects leaching rate and cycle of the uranium ore. Grain size distribution is directly related with the leaching rate, the acid consumption and other technical indicators, which is a key technical indicator of heap leaching uranium mining. Whether it is reasonable or not determines the economic benefit. Artificial detection of the grain size parameter and its distribution of uranium ore is time-consuming as well as laborious, and the test results cannot give real-time feedback. However,adopting digital image processing technology combined with modern soft measurement technology and data processing technology can improve the detection efficiency and accuracy of ore grain size parameters and reducetest cost. In a word, using image analysis technology for online real-time continuous uranium ore grain size detection and automatically obtaining ore grain size distribution are the inevitable trend of development.On-line measurement of the ore grain size parameter is realized by using digital image processing techniques, and the distribution is determined. Firstly, the camera calibration is carried on and the acquisition of ore image preprocessing including operations such as graying, denoising filtering, binarization and morphological filtering is achieved. In this paper, Zhang Zhengyou's camera calibration algorithm is adopted, and various filtering algorithms are studied. Image guided filter is applied to ore image filtering for the first time. Both Otsu method and the image segmentation algorithm of PCNN model are studied, and PCNN image segmentation algorithm based on between-class posterior maximum cross entropy criterion is proposed. Mathematical morphology method is used to optimize the segmented image. Secondly, second segmentation for the preprocessed ore images is needed. As ore images after binarization are still adhered, second segmentation is used so as to acquire simply connected grain ore images. Digital image cutting algorithm based on concave point matching is proposed to secondly segment the preprocessed image. Finally, the measuring of grain size parameters of ore and their distribution are studied. The minimum projection area of minimum circumscribed ellipse and the minimumprojection area of minimum circumscribed rectangle are used to calculate the simply connected ore grain size parameters and their grain size distribution. In order to improve the accuracy of measurement, the method of parameter measurement based on shape features is proposed to improve the accuracy of grain size parameters, and the statistical distribution chart of all grain size parameters are used to guide the heap leaching uranium mining technology and increase the leaching rate of uranium ore.