A novel personal identification method based on the fusion of multi-finger knuckleprints is proposed in this paper. The identification process can be divided into the following stages: extracting and matching the knuckleprint's line feature, fusing of the matching scores of the multi-finger knucklerprints to get the total matching score and then certifying the identity of the user. The approach was tested on a database of 98 people (1,579 Knuckleprint images). The experimental results showed the effectiveness of the method: it improved the correct recognition rate to 96.62% which is more than 6 percentages higher than the best result of using a single finger knuckleprint.