Skip to Main Content
The performance of automatic fingerprint identification system relies heavily on the quality of the captured fingerprint images. A novel method for fingerprint image quality analysis has been presented, which overcomes the shortcoming most of existing methods have, considering the correlation of each quality feature as linear and paying no attention to the clarity of local texture. In this paper, ten features are extracted from the fingerprint image and then Fuzzy Relation Classifier is trained to classify the fingerprint images, which includes the unsupervised clustering and supervised classification to care more about the revelation of the data structure than other classifiers. Experimental results show that the proposed method has a good performance in evaluating the quality of the fingerprint images.