IDENTIFIKASI TANDA TANGAN MENGGUNAKAN EKSTRAKSI FITUR ZONING DAN EUCLIDEAN DISTANCE
Daftar Isi:
- The signature is one of the biometrics owned by humans, one of the attributes most widely accepted as an identification system in recognizing someone. In this study, the author makes a system can help identify signatures in reality using a webcam. And add some results of the trial scenario. In this study the algorithm used is Euclidean Distance for recognition, and feature extraction using zoning. At the data training stage, the signature image must go through several stages of pre-processing processes such as grayscaling, thresholding, cropping, resizing, and extraction of zoning features. Then the new testing phase, using the calculation of the euclidean distance system will make decisions based on the proximity of the distance between the training data and test data. The results of system testing are those that can identify the owner of the signature. In this study used 140 samples of training data, 40 samples for non-realtime testing, and 20 samples for testing in realtime. The proposed system obtains an accuracy rate of 85% for non-realtime, and 70% for realtime results. To find out the accuracy of the algorithm used, the authors added several results of the test scenario, namely, the thickness of the pen, the light, the distance and the damaged image.