Ghost Suppression Algorithm Based on TFDT Modeling,ERICDATA高等教育知識庫
高等教育出版
熱門: Qiaoyun Zhang  Scopus AI  Helong Xia  美国 通识教育计划  Semi-trusted loader  Zhi-Wei Sun  
高等教育出版
首頁 臺灣期刊 學校系所 學協會 民間出版 大陸/海外期刊 政府機關 學校系所 學協會 民間出版 DOI查詢服務 DOI註冊服務
閱讀全文
篇名
Ghost Suppression Algorithm Based on TFDT Modeling
並列篇名
Ghost Suppression Algorithm Based on TFDT Modeling
作者 Guo-Liang Xu、Hang Zhou、Liang-You Yuan
英文摘要
In the background modeling, moving objects would often enter the scene. “Ghost” emerges as the pixels of interfering objects damage the background model. This paper presents a “ghost” suppression algorithm based on two - frame difference method and topology (TFDT). First in TFDT algorithm is to obtain the change mask with two-frame-difference method detecting the difference between two adjacent frames. Then according to the change mask, the complete object region and its bounding box are estimated with topo-logical structure. Finally, the pixels in the bounding box are eliminated, and the remaining pixels are added to the background model. TFDT is a universal algorithm for background modeling and can be adapted in foreground detection algorithms such as Codebook and ViBe. It can effectively remove object’s pixels and build a background model for foreground detection algorithms. In this work, TFDT is used to improve Codebook and ViBe, and the “ghost” is absent from the result, which has been verfied in the CDnet database.
起訖頁 053-064
關鍵詞 background modeling、two-frame difference、topological structure、foreground detection、universal algorithm
刊名 電腦學刊  
期數 202102 (32:1期)
DOI 10.3966/199115992021023201005   複製DOI
QR Code
該期刊
上一篇
An Information Security Protocol for Automotive Ethernet
該期刊
下一篇
Behaviour Classification of Cyber Attacks Using Convolutional Neural Networks

高等教育知識庫  新書優惠  教育研究月刊  Ericdata高等教育知識庫