Font Size: a A A

Study On Key Technology Of Special Steel Labeling Vision Recognition System

Posted on:2019-07-06Degree:MasterType:Thesis
Country:ChinaCandidate:Y S RenFull Text:PDF
GTID:2321330542485362Subject:Mechanical engineering
Abstract/Summary:PDF Full Text Request
Users are very strict of special steel billet and special steel round bar quality requirements(this article collectively referred to as special steel),for example: The same product must be using the special steel rolled out of the same furnace.Thence that special steel must be labeled before out the factory using to distinguish different furnace number of products.At present,the method of labeling is manual at home and abroad iron and steel enterprises and often misses labeling,caused a bad impact of iron and steel enterprises.For this reason,that high precision automatic labeling robot system of special steel end face is hotspot at the labeling field.Among them,recognizing the each special steel end face images and orientating the 3D real-time center positions based on machine vision and digital image processing technology are the preconditions of guidance robot labeling.In view of this,this thesis focuses on the extract method of special steel end face centered coordinate based on vision.This thesis is an important part of the research content of the provincial key project in cooperation with A Iron and Steel Co.The main contents and innovations of this paper are as follows: The hardware system of the special steel end face center location is established;Existence three types over-segmentation in the processes of the special steel round bar end face image segmentation,a equilateral-hexagonal-6-crossed rotation operator was proposed to identify and eliminate welding scar image on one special steel round bar end face in the round bales before the image segmentation;The small amount over-segmented image restoration is realized by "stitch bond" the boundaries;A linear proportion operator is put forward to identify and eliminate the over-segmented image of the support pedestals in the segmented round bales image;Improved SUSAN corner detection algorithm is proposed,the image plane coordinates of corners point is extracted,and the number of the special steel billets end faces image in each enclosed area is calculated.According to the closed area in the image is one or more special steel billet end face images,respectively extracting the center image plane coordinates of the special steel billet end face image;In order to make the hardware system to achieve intelligent labeling,the software design and development of special steels automatic labeling system is complete using the MATLAB development platform;The robot labeling system of the special steel round bales and the special steel billets end face was set up,and the labeling experiment was carried out respectively.Experimental results indicate that the horizontal and vertical labeling positionerror range of each special steel round bar end face is-1474.0~3222.0 and-3990.0~2940.0 respectively;horizontal and vertical labeling position error range of each special steel billet end face is-850.0~625.0 and-875.0~550.0 respectively,which meets the enterprises allowed center coordinates position error value 1 mm of the special steel end face fully.The method proposed in this paper can accurately identify the images of the special steels end face in the image,and accurately locate the center coordinates of the special steels end face,and provide the necessary coordinate information for the automatic guidance robot labeling.This would lay a theoretical and experimental foundation for the way of special steels smart labeling.
Keywords/Search Tags:Special steel robot labeling, Round steel end face, Billet end face, Machine vision, Special steel end face center location
PDF Full Text Request
Related items