Edges in image is important for object recognition and computer analysis, and a large number of edge extraction algorithms are available. This paper proposes an improved method on the basis of a half-neighborhood operator. In order to decide whether a pixel belongs to an edge, a 3-by-3 block centered at the pixel is chosen, and the maximum, minimum, mean, and standard deviation are calculated. The standard deviation is used as a threshold value, and at the same time, the ability of human eyes in distinguishing gray-level differences is taken into consideration. Edges in a number of images are extracted using the described method, and the results show validity of the method.