The invention discloses a parameter-free hand back vein maximum circular region extraction method, and belongs to the technical field of biological feature recognition. The invention aims to provide aparameter-free hand back vein maximum circular region extraction method for performing direction correction on a hand back maximum circular region so as to extract and correct the maximum circular region of the hand back region. The method comprises the following steps: (1) carrying out binarization processing on a dorsal vein image by using a conditional random field image segmentation algorithm; (2) calculating Delaunay triangulation of the coordinate sequence by using an MATLAB function delaunay and triangulation; (3) finding out a maximum radius and a corresponding circle center coordinate to obtain a maximum circular area image Cmax of the hand back; (4) selecting a contour point near the little finger, and calculating the distance between the contour point and the circle center, namely the vertex of the little finger; and (5) calculating an included angle between the OA direction and the horizontal direction. The method has the advantages that the maximum circle extraction method does not need to carry out parameters and is low in time complexity; the invention provides a method for correcting the direction of a circular area.