BROWSE
Detail
Conference Paper
객체 탐지를 위한 부가 학습: YOLO를 사용한 이중 모델 접근 방식
Cited - time in 
Share 
Share
- Issue Date
- 2024-06
- Citation
- 대한전자공학회 학술 대회 (하계) 2024, pp.2990-2994
- Publisher
- 대한전자공학회
- Language
- Korean
- Type
- Conference Paper
- Abstract
- This paper introduces enhancing object detection capabilities through an additive learning framework integrating two detection models such as YOLO (You Only Look Once). Additive learning involves AI modules collaboratively enhancing their performance by sharing and utilizing knowledge from each other. This framework significantly improves both dual models performance through continuous analysis of new incoming image data, enabling dynamic decisions on whether an image should be saved for further detection capability refinement. The implementation of this framework not only achieves higher accuracy in object detection systems but also ensures that the models adapt and improve when encountering new, unseen data points.
- KSP Keywords
- Dynamic decisions, Image data, Intrusion detection system(IDS), Learning framework, Object detection, detection capability