Fingerprint recognition system aims to identify someone. The main obstacle in recognizing a person's fingerprints is that the image has a low quality. The quality is caused by the type of oily, dry, dirty skin and the type of fingerprint scanner used. For this reason, an increase was made with the aim of improving fingerprint images as the main factor determining the level of accuracy of fingerprint image recognition results. In order for fingerprint images to be easily interpreted by humans and machines, it is necessary to improve quality by clarifying the fingerprint line. This study aims to improve the quality of fingerprint images with the Minutiae Extraction algorithm and the Learning Vector Quantization (LVQ) method as a test. This test is carried out using 4 images as training in each class. The total number of classes in this test is 10 classes with each class consisting of 10 images. In testing 75 fingerprint image data obtained an accuracy rate of 83.3333%.
Copyrights © 2020