The microscopic colors of head hair were compared using numerical values provided by image analysis of microscopic digital images of hair.   In the morphological examination of hair, color is considered to be one of the most important characteristics. In visual observation of hair color, macroscopic and microscopic observation have been carried out. “Dark brown” and “brown” are used as basic colors for the indication of the hair color. In the microscopic observation, it is common to perform more detailed observations of lightness, hue and color changes along the hair shaft. Some mastery of skills is necessary for visual inspection of hair color. Therefore, we examined a measurement method that takes hair color as digital data and an indication method that was suitable to compare hair colors.   By always checking the luminance of the background in the finder of a digital camera, it became possible to acquire hair images in a constant condition. The measurement of color was enabled with high reproducibility by arranging the measurement area to the appropriate position on images of individual hairs.   It was suggested that it was suitable to use the CIELAB color system to specify hair color. On the a^*-b^* diagram, the color of each hair and the range of lightness and reddish tinge of control hairs are shown as a coordinate point and a 95% probability ellipse, respectively, so that expressions like “light brown” or “reddish black brown” can be grasped graphically.   We evaluated a discrimination of the newly devised inspection method by calculating the exclusion ratio from the data of other persons in a personal control color domain (95% probability ellipse) using 254 head hairs collected from ten persons as samples. As a result, a high discrimination was shown depending on the control subject or on a combination of subjects to compare.