skip to main content
Ngôn ngữ:
Giới hạn tìm kiếm: Giới hạn tìm kiếm: Dạng tài nguyên Hiển thị kết quả với: Hiển thị kết quả với: Chỉ mục

基于长时间视频序列背景建模框架的运动目标检测方法
Moving object detection method of background modeling frame based on long-time video sequence

Ding Jie ; Xiao Jiangjian ; Song Kangkang ; Peng Chengbin

Truy cập trực tuyến

Trích dẫn Trích dẫn bởi
  • 基于长时间视频序列背景建模框架的运动目标检测方法

  • Nhan đề:
    Moving object detection method of background modeling frame based on long-time video sequence
  • Tác giả: Ding Jie ; Xiao Jiangjian ; Song Kangkang ; Peng Chengbin
  • Chủ đề: Physics ; Computing ; Calculating ; Counting ; Recognition Of Data ; Presentation Of Data ; Record Carriers ; Handling Record Carriers ; Medicine ; Sciences ; Chemistry ; Physics
  • Mô tả: The invention provides a moving object detection method of a background modeling frame based on a long-time video sequence, which comprises the following steps: (a) acquiring a long-time panoramic video by a fixed panoramic camera; (b) decoding the long-time panoramic video and reasonably describing the processed long-time panoramic video through a stack-type self-coding neural network to obtain aplurality of background descriptors; (c) organizing the background descriptors by clustering to obtain typical background descriptors, and organizing the typical background descriptors into a hierarchical tree to obtain a background word bag; (d) using a ViBe background modeling method to fuse the background word bag to obtain a background model framework so as to obtain a background model, and then detecting the moving object through the background model. 本发明提供种基于长时间视频序列背景建模框架的运动目标检测方法,其包括如下步骤:(a)通过固定式全景摄像机获取长时间全景视频;(b)对所述长时间全景视频进行解码处理,并通过栈式自编码神经网络对处理后的长时间全景视频进行合理描述,得到多个背景描述子;(c)通过聚类组织所述背景描述子,获取典型背景描述子,并将典型背景描述子以层次树的形式进行组织得到背景词袋;(d)采用ViBe背景建模方法对该背景词袋进行融合处理,得到背景建模框架,进而获得背景模型,再通过该背景模型对运动目标进行检测。
  • Năm xuất bản: 13 November 2018
  • Ngôn ngữ: Chinese;English

Đang tìm Cơ sở dữ liệu bên ngoài...