Orientation consistency based feature extraction for fingerprint identification

Feature extraction is one of the key techniques in automatic fingerprint identification. Today, most of feature extractions are based on the structure analysis of fingerprint ridges and valleys. The structure analysis depends on the accuracy of image segmentation. Image segmentation is one of the most difficult techniques in image processing. Therefore, the accuracy and correctness of the feature extraction are often not reliable. For this reason, we propose a method for feature extraction in fingerprint identification. In this method, the intensity gradient vector image is first generated using expanded Prewitt operator. Then the orientation consistency of the gradient vector image is calculated. The feature points of the fingerprints are extracted from the orientation consistency by classification. This method is fast, simple and accurate.