<![CDATA[The Electronic Identity Card (e-KTP) serves as the official proof of identity for residents, issued by the relevant implementing agency across the entire territory of the Unitary State of the Republic of Indonesia. Mandatory for both Indonesian citizens (WNI) and foreigners (WNA) holding a Permanent Stay Permit (ITAP) and aged 17 or married, the e-KTP is susceptible to potential damage, often arising from factors such as prolonged usage or improper handling. Physical damage to the e-KTP can impede the document's ability to accurately verify identity, potentially impacting public services and government administration. This research aims to assess the condition of e-KTPs, determining whether they are in good or damaged condition. The study employs the Convolutional Neural Network (CNN) method, known for its significant results in image recognition by attempting to emulate the image recognition system in the human visual cortex, facilitating the processing of image information. This method comprises two architectural layers: Feature Learning and Classification. The dataset utilized in this research comprises images of e-KTPs sourced from the Population and Civil Registration Office of Bengkalis Regency, totaling 400 images categorized into two classes: 200 for good condition and 200 for damaged condition. The research findings enable the determination of the e-KTP image's condition, achieving a 90% accuracy rate.]]>
Copyrights © 2024
