BACK-END SUBSYSTEM DEVELOPMENT FOR CHAOS MAP GENERATION AND DATABASE ACCESS FOR VEHICLE REGISTRATION CERTIFICATE EXTENSION APPLICATION WITH IMAGE ENCRYPTION

• Main Author: Mulya Harjono, Hafizh
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/55956
• Institution: Institut Teknologi Bandung

The Vehicle Registration Certificate is used as proof of ownership and proof of the legality of operating motorized vehicles in Indonesia, so every vehicle owner is required to have it. To extend the certificate, the owners need to come directly to Samsat with the required documents. However, the extension process at Samsat is not efficient because most of the time is done by queuing and waiting. The extension process can be simplified by using a mobile application as it makes it possible to do the process from home. This scenario needs special attention to maintain data confidentiality and security of the uploaded documents that contain sensitive data. This project proposed a backend API subsystem for image encryption to provide chaos maps and storage services to securely upload and store documents for the Vehicle Registration Certificate extension by using a mobile app. The proposed system helped to make the Vehicle Registration Certificate extension more effectively done because it enabled vehicle owners to do it from their homes. This project tried to approach the problem by using Henon’s map and Arnold’s cat map algorithm to encrypt image documents and using a database service to store the documents. Arnold's cat map has a limitation in that it can only accept square images. This project uses a NoSQL-based external database service to store the documents. The result showed that for image size between 50 pixels x 50 pixels to 1000 pixels x 10000 pixels, Arnold’s cat map algorithm’s API response time grew exponentially to image size, while the Henon’s map’s response time grew significantly slower with a correlation coefficient of 0.368 (weak correlation). Arnold’s cat map’s response time also grew proportionally to the number of iterations that were used as the secret key, so there was a trade-off between security and response time. This project concluded and served as a proof-of-concept that Henon’s map algorithm could be used to encrypt image-based documents, while Arnold’s cat map algorithm, which had a slow response time, was not feasible to be implemented in a mobile app environment.