Research On Ore Particle Size Detection Method Based On Image Processing

The ore particle size is an important index of grinding operation technology.Grinding operations need to detect the particle size of ore,and adjusting the process parameters according to the test results.At present,the conventional methods such as screening and sedimentation are used to measure the particle size,but these detection methods are time-consuming and inefficient,and the test results are easily affected by the subjective impact of the testing staff.Aiming at the deficiency of the previous measurement methods,this paper uses the detection method based on digital image processing.It is proved by practice that this detection method can quickly and accurately analyze the measurement of the particle size of ore.The main contents of this paper:1.Pretreating the original ore sample image.Firstly translated the original image into gray image,next used the contrast adjustment aiming at the phenomenon that ore particle and background are not easy to distinguish.Then analysed and compared two typical filtering arithmetics and choose the median filter algorithm to remove the image noise.2.Useing the segmentation arithmetic on the ore sample image.This paper analyzes and compares different segmentation.Finally choosing the images segmentation method which is based on threshold and separate ore particle from background successfully.3.Morphological processing on the ore sample image.This paper used morphological operations on the segmented image and smoothed the image noise,filling the hole which is formed by the reflection of the ore particle.4.The segmentation of adherent ore particles.On the basis of the traditional watershed algorithm,this paper analyzes and compares several improved watershed segmentation results.Finally,the paper used watershed segmentation algorithm which is based on the marker control to separate the ore particle.5.Labeling the connected regions in the image,then calculate the number of pixels in each connected regions,get the actual ore particle size by the proportion of conversion.6.Completed the software compilation and output the size distribution curve,compared the results of the software with the screening results.Proved by experiments,this measurement method detects the ore particle size distribution successfully and effectively.Compared with traditional methods,the detection method which is based on image processing has the advantages of simple operation,high accuracy and high efficiency.At the same time,the on-line detection of ore particle size has also achieved good results.This method has great significance to improve the efficiency of grinding operation and promote the development of automation of mineral processing.