<![CDATA[Character education in elementary schools is still dominated by lecture methods, so students are less interested and have difficulty understanding the material optimally. This study aims to develop computer vision- based character education media with a markerless approach using the FAST Corner Detection algorithm to improve the understanding of fourth grade students. The research method used is the Multimedia Development Life Cycle (MDLC) which includes the stages of concept, design, material collecting, assembly, testing, and distribution . The application was developed using Unity and Vuforia on the Android platform, then tested through Black Box Testing , distance, angle, light intensity testing, User Acceptance Testing (UAT), USE Questionnaire , and pre-test and post-test analysis . The results showed that all application functions ran well. The FAST Corner Detection algorithm was able to detect images optimally at light intensities of 11–20 lux to 26,378–29,413 lux. The average pre-test score increased from 73.75 to 91.25 in the post-test , indicating an increase in student understanding after using the application. The UAT and USE Questionnaire results also demonstrated excellent acceptance and user experience. Therefore, the developed learning media is suitable for use as a more interactive, engaging, and effective alternative for character education for fourth-grade students. Character education in elementary schools is still dominated by lecture methods, so students are less interested and have difficulty understanding the material optimally. This study aims to develop computer vision-based character education media with a markerless approach using the FAST Corner Detection algorithm to improve the understanding of fourth-grade students. The research method used is the Multimedia Development Life Cycle (MDLC) which includes the stages of concept, design, material collecting, assembly, testing, and distribution. The application was developed using Unity and Vuforia on the Android platform, then tested through Black Box Testing, distance, angle, light intensity testing, User Acceptance Testing (UAT), USE Questionnaire, and pre-test and post-test analysis. The results showed that all application functions ran well. The FAST Corner Detection algorithm was able to detect images optimally at light intensities of 11–20 lux to 26,378–29,413 lux. The average pre-test score increased from 73.75 to 91.25 in the post-test, indicating an increase in student understanding after using the application. The UAT and USE Questionnaire results also demonstrated excellent acceptance and user experience. Therefore, the developed learning media is suitable for use as a more interactive, engaging, and effective alternative for character education for fourth-grade students.]]>
Copyrights © 2026
