Research On Technique Of Recognition And Location For Circular Coded Diagonal Target

The artificial cooperation target can assist in the measurement process and improve the accuracy and reliability of the measurement.Among them,the coded target is different from the general cooperation target,which can realize automatic matching according to its own identity information,so it has greater potential to be developed and exploited.Many scholars have designed a variety of coded targets,and more commonly used now are circular coded targets.After a great deal of research,a kind of circular coded diagonal target(CCDT)is selected as the research object in this paper,and the advantage of this coded target is analyzed.The identification and location algorithm is studied in this paper.First,a series of image preprocessing method is used to process the circular coded diagonal target,and then,OTCircles algorithm and circle frequency filter algorithm are used to detect the CCDT to obtain the coarse positioning position.Here,Based on this,this paper focuses on the sub-pixel location method of diagonal coding.Finally,the fast decoding method for CCDT is realized based on the coordinate transformation.This article mainly completed the following three aspects of work:(1)In the process of circular coded diagonal target detection algorithm,aiming at solve the problem of uneven illumination in the image,an improved adaptive Gamma correction algorithm is adopted,which has better robustness than the existing target detection algorithm in dealing with different lighting conditions.This paper makes full use of the characteristics of this research object.For the detection of the center circle of the code diagonal target,the OTCircles detection algorithm is adopted,and the ellipse detection is adjusted by adjusting the parameters to realize the efficient detection of the ellipse.Combined with the circle frequency filtering algorithm,using the inner circle diagonal,for its rules on changes in the gray values,achieve a simple circle diagonal detection;combined with these two kinds of features,the use of two algorithms,the paper achieved a high efficiency and high accuracy detection of circular coded diagonal target.After the coarse positioning is completed,the criterion is set in this paper to detect whether the CCDT is severely distorted.For those that affect the decoding accuracy,this paper uses the perspective correction algorithm to correct the CCDT image.(2)This paper attempts three different positioning methods in the precise location of the CCDT.Again,based on the design characteristics of the CCDT,its precise positioning can make use of both the center circle and the circular diagonal in the center circle.For the center circle,this paper adopts the sub-pixel positioning algorithm of surface fitting.For the diagonal of the circle,this paper adopts the gradient sub-pixel corner detection algorithm and the sub-pixel positioning algorithm of the correlation coefficient fitting extreme value.A series of diagonal emblem images were coded to compare and analyze the performance of the three precise positioning algorithms.The algorithm with high precision,relative robustness and complexity was selected.And finally compared with the commercial software,the real experiment results and the simulation results are analyzed.The algorithm in this paper is better than the commercial software in recognition rate and positioning accuracy.(3)Introduced the basic idea of decoding the CCDT,and realized the decoding algorithm based on the coordinate transformation.