This paper proposed a fingerprint matching algorithm based on the bifurcation ridge comparability, which found pairs of possible reference minutiae according to the comparability of two ridges on which the highly reliable bifurcation lied. Meanwhile, in order to solve the problem that the reference minutiae filtering would be influenced by noise, a novel method is proposed taking advantage of the relationship between the reference minutiae and the subsets consisting of four neighboring minutiae to determine the final reference minutiae and the transform parameters. And finally the algorithm implemented the fingerprint matching using the variable bounding box. The experiment results indicated that this algorithm could accurately and quickly locate the reference minutiae and precisely acquire the transform parameters, then correctly and quickly perform the fingerprint matching.