Penentuan Karakter Anak Melalui Sidik Jari dengan Algoritma Convolutional Neural Network (CNN)

Abstract

Each individual has a character that is influenced by internal and external factors. A deeper understanding of the character and potential of the child can help parents optimally direct their development. Dermatoglyphics is a study that examines the protrusions or patterns of the epidermis on the fingers and palms formed since the development of the fetus' nerves in the womb. This study is considered to be able to analyze the character and potential of the child. In addition, the presence of the IGF2R gene (Insulin-Like Growth Factor 2 Receptor) can affect a person's intelligence and character. On the other hand, along with the development of this digital era, everything has become more practical. One of them is fingerprints that are digitally identified to determine the character and potential. The Convolutional Neural Network (CNN) algorithm is one method that can be used to analyze a child's character through fingerprints. The data used is 300 data with the data divided into 210 training data, 60 validation data, and 30 testing data. The data will go through several pre-processing stages, namely labelling, greyscale, resizing, cropping, normalization, and augmentation. Based on the tests that have been carried out, the system can analyze fingerprint images with an accuracy value of 94%. The model that has been created is integrated into the form of a website.