Please use this identifier to cite or link to this item: http://drsr.daiict.ac.in//handle/123456789/1193
Title: Content-Based Video Retrieval Based on Integration of Wavelet Transform, Color and Texture Features
Authors: Khare, Manish
Thakkar, Jahanvi
Keywords: Video retrieval
Discrete Wavelet Transform
Edge Histogram Descriptor
Issue Date: 2023
Publisher: Dhirubhai Ambani Institute of Information and Communication Technology
Citation: Thakkar, Jahanvi (2023). Content-Based Video Retrieval Based on Integration of Wavelet Transform, Color and Texture Features. Dhirubhai Ambani Institute of Information and Communication Technology. vii, 40 p. (Acc. # T01134).
Abstract: High-resolution, large-sized videos can now be transferred due to the fast developmentof information and communication technology, and video applicationshave developed in line with data quality levels. The applications of contentbasedvideo retrieval (CBVR) in various fields, like surveillance, education, sports,medicine etc., make it a crucial video application. The efficient Content-BasedVideo Retrieval (CBVR) algorithm described in this thesis is based on the MPEG-7 features of the Discrete Wavelet Transform (DWT), Dual-Tree Complex WaveletTransform (DTCWT), Edge Histogram Descriptor (EHD), Linear Binary Pattern(LBP) and colour Layout Descriptor (CLD).This thesis proposes a content-based video retrieval system that integrates wavelettransform, colour, and texture features for efficient and accurate video retrieval.The proposed system aims to address the limitations of traditional video retrievalsystems that rely solely on low-level features, such as colour and texture, and donot consider the video�s structural information. The system utilizes the wavelettransform to extract the structural information of the video. It combines it withcolour and texture features to create a robust and accurate feature set for retrieval.
URI: http://drsr.daiict.ac.in//handle/123456789/1193
Appears in Collections:M Tech Dissertations

Files in This Item:
File SizeFormat 
202111058.pdf9.59 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.