Utilizing ORB Algorithm in Web-Based Sales Application
| Autor: | Edward Brainard Pranata, Tony Tony |
|---|---|
| Jazyk: | English<br />Indonesian |
| Rok vydání: | 2024 |
| Předmět: | |
| Zdroj: | Journal of Information Systems and Informatics, Vol 6, Iss 1, Pp 378-398 (2024) |
| Druh dokumentu: | article |
| ISSN: | 2656-5935 2656-4882 |
| DOI: | 10.51519/journalisi.v6i1.671 |
| Popis: | E-commerce has become common and important for businesses, but Jaya Sentosa Store has not implemented it. E-commerce commonly has only a search by keyword feature, but that cannot replicate Jaya Sentosa Store order process. An image-based search is needed to replicate the order process. Our research purpose is to develop a web-based sales application and an image search feature for Jaya Sentosa Store. We apply Scrum when developing this application. We use Javascript (JS) programming language. Back-end and front-end development employ Express JS and React JS framework, respectively. To get the right feature-matching algorithm, we conduct a test between the SIFT, KAZE, and ORB algorithms. We write Python scripts to implement ORB algorithm in image-based search feature. Our test shows that the ORB algorithm has the fastest average running time, i.e., 3.415 s, compared to SIFT and KAZE. Black box testing of the sales application shows that all cases are valid. It means that our application can replicate Jaya Sentosa Store order process and gain a competitive advantage. |
| Databáze: | Directory of Open Access Journals |
| Externí odkaz: |
|