This paper proposes a self-localization method for automated vehicles by using traffic signs. The proposed method aims to improve the accuracy of longitudinal self-localization. In order to search for the highest probability position, map matching is performed using predefined digital map data. The images for matching, specifically, (Filtered sign image, Ideal sign image) are made at each position in longitudinal direction. A filtered sign image is extracted from the road forward camera image whereas an ideal sign image is created using the digital map data. The similarity between these two images is then calculated using ZNCC (Zero-means Normalized Cross Correlation). The corresponding posterior probability of the similarity is updated by the BBF (Binary Bayes Filter) and the vehicle position is estimated accordingly. The actual experiments were performed based on two traveling data. The result obtained using the proposed method indicated that the localization accuracy in the longitudinal direction was improved by approximately 15.0% regarding RMS (Root Mean Square) error.