As a global feature of fingerprints, the orientation field plays an important role in automatic fingerprint identification systems. Although many algorithms have been proposed for orientation field estimation, the results are not so satisfactory and the computational cost is expensive. In this work, a novel algorithm based on a straight-line model of ridge is proposed for the orientation field estimation. The algorithm comprises four steps: preprocessing the original fingerprint image, determining the primary and secondary ridges of the fingerprint foreground block using the top semineighbor searching algorithm, estimating the block direction based on a straight-line model of such a primary ridge, and correcting the spurious block directions. The proposed algorithm is suitable for all types of fingerprints. Experimental results show that it achieves satisfying estimation accuracy with high computational efficiency. A further experiment shows that it is more accurate and robust to noise compared with previous works, and can improve the performance of a fingerprint recognition system.