1 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 1 / 1
Öğe Fingerprint recognition system based on gray level co-occurrence matrix(INESEC, 2017) Çalışkan, AbidinThe biometric system provides an automatic identification of any person, depending on characteristic and feature/attribution of person. Fingerprint is, today, one of the biometric systems that have a wide range of use in many investigation areas. Fingerprint, especially used for authentication, is more reliable comparing to the other traditional methods which are used for access. In this study, a gray level co-occurrence matrix (GLCM) based fingerprint recognition system which provides successful results in tissue type imagining recognition has been implemented. The purpose of this study is to show the effectiveness of the GLCM in fingerprint recognition. By using GLCM which is a feature extracting method, fingerprint images are classified by multilayer perceptron (MLP) artificial neural network classification technique. Statistical methods were used to extraction the feature by obtaining the GLCM matrix for the gray level images. In the first step of system analysis, the system is trained by using GLCM attribute parameters and performance information is measured for different network topologies of the MLP classifier. After the classification stage, when the results are compared with the success rates of previously made fingerprint recognition systems, the success rate of 88.25% is considered as acceptable. As a result, it is considered that the results are reasonable when results are compared with other studies in the literature. Experimental results have also shown that the proposed method can improve the accuracy of existing methods.
