퓨샷 학습과 삼중항 손실을 이용한 지상 무기체계 분류 모델의 확장성 연구
Research on Extensibility of Ground Weapon System Classification Model Using Few-shot Learning and Triplet Loss
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This paper proposes an extensible ground weapon system classification model based on a dataset with 11 categories obtained through web crawling and publicly available operation and training videos. A base model is trained using this dataset to extract discriminative features, similar to conventional supervised learning. By applying few-shot learning and triplet loss, an extended model is trained based on the previous training. The evaluation shows the base model achieved 93.35% to 97.3% accuracy for 5 categories and 93.54% to 99% accuracy for 11 categories. The extended model, achieved 72.18% to 75.54% accuracy when trained on 5 categories and tested on 11. These experiments demonstrate the effectiveness of few shot learning and triplet loss in creating an extended model capable of predicting new, untrained categories even with limited data.
Abstract
This paper proposes an extensible ground weapon system classification model based on a dataset with 11 categories obtained through web crawling and publicly available operation and training videos. A base model is trained using this dataset to extract discriminative features, similar to conventional supervised learning. By applying few-shot learning and triplet loss, an extended model is trained based on the previous training. The evaluation shows the base model achieved 93.35% to 97.3% accuracy for 5 categories and 93.54% to 99% accuracy for 11 categories. The extended model, achieved 72.18% to 75.54% accuracy when trained on 5 categories and tested on 11. These experiments demonstrate the effectiveness of few shot learning and triplet loss in creating an extended model capable of predicting new, untrained categories even with limited data.
참고문헌 (18)
- 양동원, "형태특징과 지역특징 융합기법을 활용한열영상 기반의 차량 분류 방법", 전기전자학회논문지, 24(1), pp.97~105 (2020), 한국전기전자학회
- 김종환, "전이학습 기반 무인수색차량의 표적 탐지 및 식별 딥러닝 모델 분석", 한국국방경영분석학회지, 47(2), pp.26~35 (2021), 한국국방경영분석학회[DOI]
- 임승균, "YOLO 알고리즘을 이용한 전차 국적 식별 및 평가", 정보과학회 컴퓨팅의 실제 논문지, 27(12), pp.555~562 (2021), 한국정보과학회[DOI]
- 황진용, "기동무기체계 분류를 위한 전이학습 기반 앙상블 모델 연구", 한국정보기술학회논문지, 19(12), pp.1~10 (2021), 한국정보기술학회[DOI]
- 강세혁, "퓨샷 러닝을 이용한 미식별 전차 식별 방안 연구", 한국통신학회 학술대회논문집, pp.392~393 (2022), 한국통신학회
- 류제민, "초해상화 기반 CNN을 이용한 군사용 SAR 자동표적인식 모델 연구", 제어.로봇.시스템학회 논문지, 28(1), pp.22~27 (2022), 제어·로봇·시스템학회[DOI]
- 황교성, "데이터증강 모델 조합을 활용한 적성 전차에 대한 객체탐지 성능 향상 연구", 한국CDE학회 논문집, 27(2), pp.148~159 (2022), 한국CDE학회
- 이영균, "포병 무기체계 분류를 위한 CNN 모델의 성능 비교 연구", 한국산학기술학회논문지, 24(1), pp.344~350 (2023), 한국산학기술학회[DOI]
- 황재민, "SinGAN을 활용한 군사장애물 데이터 부족 문제 해결 및 실증 연구", 한국산학기술학회논문지, 24(1), pp.310~317 (2023), 한국산학기술학회[DOI]
- Bromley, J., "Signature verification using a" siamese" time delay neural network", Advances in neural information processing systems, 6, pp.737~744 (1993)
- Hadsell, Raia, "Dimensionality reduction by learning an invariant mapping", 2006 IEEE computer society conference on computer vision and pattern recognition (CVPR'06), 2, pp.1735~1742 (2006), IEEE
- Lake, B., "One shot learning of simple visual concepts", Proceedings of the annual meeting of the cognitive science society, 33(33), pp.2568~2573 (2011)
- Schroff, F., "Facenet: A unified embedding for face recognition and clustering", Proceedings of the IEEE conference on computer vision and pattern recognition, pp.815~823 (2015)
- Finn, C., "Model-agnostic meta-learning for fast adaptation of deep networks", International conference on machine learning, pp.1126~1135 (2017)
- Xiaozhu, X., "Object detection of armored vehicles based on deep learning in battlefield environment", 2017 4th International Conference on Information Science and Control Engineering (ICISCE), pp.1568~1570 (2017)
- Liu, Y., "Classification of tank images using convolutional neural network", 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), pp.210~214 (2020)
- Zhang, L., "Azimuth-Aware Discriminative Representation Learning for Semi-Supervised Few-Shot SAR Vehicle Recognition", Remote Sensing, 15(2), pp.331~ (2023)
- "Keras Applications" (2023)[URL]