The Charpy impact test is used to identify the transition between ductility and brittleness. The percentages of ductile and brittle fractures in steel can be evaluated based on each fracture area, which is presently determined by an analyzer with the naked eye. This method may lead to subjective judgement, and difficulty accurately quantifying the percentage. To resolve this problem, a new analysis method based on image processing is proposed in this study. A program that can automatically calculate the percentage of the ductile and brittle fractures has been developed. The analysis is performed after converting an RGB fracture image into a binary image using image processing techniques. The final binary image consists of 0 and 1 pixels. The parts with the pixel values of 1 correspond to the brittle fracture areas, and the pixel values of 0 represent the ductile fracture areas. As a result, by counting the number of 0 pixels in the entire area, it is possible to automatically calculate the percentage of ductile fracture. Using the proposed automatic fracture analysis program, it is possible to selectively distinguish only the brittle fracture from the entire fracture area, and to accurately and quantitatively calculate the percentages of ductile and brittle fractures.