| With the improvement of request on quality of round bale from industrial production,labeling on the round bale end faces has become an indispensable process before delivery.At present,the method of labeling on the round bale end faces is manual at home and abroad.Manual labeling is labour-intensive,inefficient and low accuracy,it can’t meet the labeling requirements of large-scale and high precision,therefore,automatic labeling system on the round bale end faces is urgently needed in the industry.Among them,recognizing the round bale end face images,orientating the 3D real-time center positions and obtaining the required labeling pose information based on machine vision and digital image processing technology are the preconditions of executing labeling operation.In view of this,this paper focuses on the technologies of end face images recognition and its center location of round bales.This paper is an important part of the research content of the provincial key project in cooperation with Shijiazhuang iron and Steel Co.The main contents and innovations of this paper are as follows:(1)The general architecture of labeling system is designed based on the labeling principle of the round bale end faces,and the working principle,hardware selection and system construction of machine vision recognition subsystem are deeply studied.(2)According to the imaging characteristics of round bale end face,a whole set of image recognition method is proposed.First of all,gray image,image binarization,image filtering and other algorithms are used to pre-process the image,then the watershed algorithm is used to segment the adhesion image.Meanwhile,in order to recognize the labeling images and orientate the centers of round bale end faces,true roundness and angle threshold combined with edge detection and ellipse fitting method are carried out.(3)The conversion relationship between the pixel coordinates and world coordinates is completed by the calibration based on the Interpolation of Delaunay Triangles.(4)The labeling experimental system is set up,and a labeling experiment was conducted to validate the reliability and feasibility of study contents in this paper on the aspects of labeling speed and effect.The whole set of method proposed by this paper can recognize round bale end faces image accurately and locate the center positions precisely,and it provides pose information for labeling system.Experimental results show that,the whole set of method can meet the labeling system accuracy requirement,furthermore,it can provide theoreticalreference and technical support for the automatical and high efficiency labeling of round bale end faces. |