An approach is proposed to make the extraction and recognition of circ ular coded target (CCT) more efficient and robust.Characteristics and the encoding scheme of CCT are introdu ced at first.Then,the extraction of CCT is accomplished through three steps as follows:approximate localizat ion of the imaging center using a proposed 6-points algorithm;edge detection;ellipse fitting and edge screening .Subsequently,the recognition of CCT is realized by a presented ellipse-incircle-division algorithm based on th e invariance to the affine transformation of CCT.Finally,experiments are performed and the results are d escribed.The extraction algorithm proposed is efficient in the whole,and it can reject the di sturbance effectively with less elapsed time.The maximum and the average errors of ellipse fitting are less than 0.06p ixel and 0.03pixel,respectively, which suggests its high accuracy.In addition,the CCT recognition algorithm pre sented is quite robust,and it is able to generate desirable code results in extreme conditions.When applied to t he images containing complicated background information,the global omissive,false recognition rates and the ave rage elapsed time are 4.34%, 0.47% and 7.22ms,respectively,which validates the optimality of its comprehens ive performance compared with other existing recognition algorithms.