In an increasingly developed digital era, the need for efficient and accurate document management is rising. One of the solutions that can be used is a scanner application based on Optical Character Recognition (OCR) algorithms, which can convert text in images to editable digital data. This research aims to design a scanner application using OCR with an Artificial Intelligence (AI) approach to improve the accuracy of text extraction results. This application is built using React Native as the interface framework, Supabase as the backend, and AI integration through OpenAI. Testing shows that the application works well, with a success rate of 95% without any fatal errors. Meanwhile, 5% of the tests passed with minor notes related to the quality of the input image. Research prove that this application can be a practical solution in supporting document digitization, especially in campus environments and other institutions. With user-friendly features and reliable performance, this application has great potential for widespread use.
Huliman et al. (Wed,) studied this question.