OPTIMIZATION OF ARTIFICIAL NEURAL NETWORKS USING ANT COLONY OPTIMIZATION TO IDENTIFY SIGNATURE IMAGES

Sari

Biometrics can be used in identification or recognition systems because it is a method for recognizing humans based on one or more physical features or unique behaviors, one of which is the Signature. In its application, signatures need to be examined (identification) because signatures are often imitated or falsified for various purposes. The signature that will be identified will be taken first by its characteristics by performing the feature extraction process using the Gabor Wavelet Transform (TGW) method. After extracting, the signature identification process is carried out using the Neural Network. In the process of applying the Neural Network method, optimization will be performed using Ant Colony Optimization. The results showed that the identification of signatures made using the Neural Network alone produced an accuracy of 77%, but after optimization using Ant Colony Optimization increased to 83%.