열적외선 영상을 사용한 미상 객체 검출 시스템
Target-agnostic Object Detectionusing Infrared Thermal imagery
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
최근 열적외선 카메라는 군사, 농업, 감시 및 의료 목적으로 다양한 응용 분야에서 활용되고 있다. 그러나 열적외선 카메라의 높은 가격과 부족한 공개 데이터 세트로 인해 딥러닝 기반 객체 탐지 알고리즘을 개발은 여전히 어려운 과제이다. 본 논문에서는 학습 데이터 없이 열 신호를 기반으로 미상 객체를 탐지하는 방법을 제시한다. 열원을 가진 객체는 주변 환경에 비해 상대적으로 높은 밝기를 가진다는 가정에 기반한 열적외선 이미지의 처리를 수행한 후, 제로 샷 분할(Zero-Shot Segmentation) 방법으로 관심 객체를 강건하게 탐지한다. 제안한 방법은 원격 제어 전차 모델과 열적외선 카메라를 사용하여 다양한 환경에서 취득한 데이터 세트로 성능을 평가하였다. 실험을 통해 제안된 방법이 열원을 가진 미상 객체를 탐지하는 데 효과적임을 보여준다.
Abstract
Recently, thermal infrared cameras are utilized for various applications, such as military, agriculture, surveillance, and medical purposes. However, it is not easy to develop deep learning-based target object detection algorithms due to high price of thermal IR cameras and insufficient public databases. In this paper, we present a method to detect target-agnostic objects based on thermal signals without any training data. After image processing of an input IR image based on the assumption that the target object has relatively higher brightness than environment, the recent zero-shot segmentation method is applied to detect the target object robustly. The dataset for target-agnostic object detection is generated in the various environments using remote control tank models and a thermal IR camera. By experiments, we show the effectiveness of the proposed method to detect target-agnostic objects.
참고문헌 (26)
- Girshick, R, "Fast r-cnn", Proceedings of the IEEE International Conference on Computer Vision, pp.1440~1448 (2015)
- Ren, S., "Faster r-cnn : Towards real-time object detection with region proposal networks", Advances in Neural Information Processing Systems, 28, pp.91~99 (2015)
- He, K., "Mask r-cnn", Proceedings of the IEEE International Conference on Computer Vision, pp.2961~2969 (2017)
- Redmon, J, "You only look once : Unified, realtime object detection", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.779~788 (2016)
- Redmon, J., "YOLO9000 : Better, faster, stronger", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.7263~7271 (2017)
- Wang, C. -Y., "YOLOv7 : Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors", Proceedings of the IEEE, 2023, 7464~7475
- Carion, N., "End-to-end object detection with transformers", European Conference on Computer Vision, pp.213~229 (2020)
- Deng, J., "ImageNet: A large-scale hierarchical image database", IEEE Conference on Computer Vision and Pattern Recognition, 2009(1), pp.248~255 (2009)
- Lin, T. -Y., "Microsoft coco : Common objects in context", Computer Vision–ECCV 2014: 13th European Conference, pp.740~755 (2014)
- Sun, C., "Revisiting unreasonable effectiveness of data in deep learning era", Proceedings of the IEEE International Conference on Computer Vision, pp.843~852 (2017)
- Srinivasan, K., "WIT : Wikipedia-based image text dataset for multimodal multilingual machine learning", Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, pp.2443~2449 (2021)
- He, L., "Fast connected-component labeling", Pattern Recognition, 42(9), pp.1977~1987 (2009)
- Kirillov, A., "Segment anything", Proceedings of the IEEE, 2023, 1235~1244
- Everingham, M., "The Pascal Visual Object Classes(VOC)Challenge : A Retrospective", International Journal of Computer Vision, 111(1), pp.98~136 (2015)
- Brown, T., "Language models are few-shot learners", Advances in Neural Information Processing Systems, 33, pp.1877~1901 (2020)
- Brown, T. B., "Language models are few-shot learners", Advances in Neural Information Processing Systems (NeurIPS), 33, pp.1877~1901 (2020)
- Improving language understanding by generative pre-training, "Radford, A.", OpenAI, ~ (2018)
- Xian, Y., "Zero-shot learning—a comprehensive evaluation of the good, the bad and the ugly", IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(9), pp.2251~2265 (2018)
- Caron, M., "Emerging properties in self-supervised vision transformers", Proceedings of the IEEE, 2021, 9650~9660
- Radford, A., "Learning transferable visual models from natural language supervision", International Conference on Machine Learning, pp.8748~8763 (2021), PMLR
- Otsu, N., "A Threshold Selection Method from Gray-Level Histograms", IEEE Transactions on Systems, Man, and Cybernetics, 9(1), pp.62~66 (1979)