The Research Of Reinforced Counting Method Based On Matlab Image Processing

Currently,the sales model of most domestic steel mills is fixed bundle branch sales,counting the number of steel is particularly important.Now the methods of many companies are manual counting,which has low efficiency,high requirements for workers,and has been unable to achieve the rapid production requirements of the modern enterprise.So,how to count steel automatic has become a problem to be solved.This paper use digital image processing approach to identify the bundle of steel digital image.At the beginning reinforced pretreatment operation to the digital image,the color image is transformed into a grayscale image,and histogram equalization,strengthen the image contrast,and filter the image;Then divided image,This paper use multiple sub-image threshold segmentation method.In order to solve the question of the target to be identify in image is not separated from each other obvious,we use a mathematical morphology algorithm of fill holes,corrosion and remove the small area,most target to be recognized has divided from each other and also removes the majority of the background and noise;Finally,use labeling method of connected component,scanned image in accordance with the regional guidelines of eight communicating,extracts the communication field area and other related parameters,use area and the area ratio counting method counts connected region so as to achieve the purpose of counting a bundle of steel.Based on pattern recognition,digital image processing technology,this paper design a reinforced counting system based on MATLAB platform.Identification and counting the steel by analysis and processing reinforced image.