DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Choi, Ho-Jin | - |
dc.contributor.advisor | 최호진 | - |
dc.contributor.author | Li, Zhun | - |
dc.date.accessioned | 2019-09-04T02:48:05Z | - |
dc.date.available | 2019-09-04T02:48:05Z | - |
dc.date.issued | 2019 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=843553&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/267110 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전산학부, 2019.2,[v, 35 p. :] | - |
dc.description.abstract | Rainfall depth is important meteorological information. Generally, high spatial resolution rainfall data such as road-level rainfall data are more beneficial. However, it is expensive to set up sufficient Automatic Weather Systems to get the road-level rainfall data. In this paper, we proposed to use deep learning to recognize rainfall depth from road surveillance videos. To achieve this goal, on the one hand, we collected new video data sets and proposed an estimation procedure to calculate refined rainfall depth from the original meteorological data. On the other hand, we proposed a new deep learning architecture named Temporal and Spatial Segment Networks (TSSN) for rainfall depth recognition. Under the TSSN framework, we utilized the video frame, the differential frame, as well as the optical flow image for rainfall depth recognition. The experimental results show that the combination of the video frame and the differential frame is a superior solution for rainfall depth recognition. The comparative experiments show that TSSN significantly outperforms other adopted methods and has higher robustness against camera movement. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Deep learning▼arainfall depth▼asurveillance video▼atemporal segment networks▼atemporal and spatial segment networks | - |
dc.subject | 심층 신경망▼a강우량▼a감시 카메라 영상▼a시간 분할 네트워크▼a시간 및 공간 분할 네트워크 | - |
dc.title | Rainfall depth recognition from road surveillance videos using deep learning | - |
dc.title.alternative | 딥러닝을 이용한 도로 감시 카메라 영상의 강우량 인식 방법 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전산학부, | - |
dc.contributor.alternativeauthor | 리 준 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.