Color Difference Histogram for Feature Extraction in Video Retrieval
Journal Title: International Journal of Computer Science & Engineering Technology - Year 2015, Vol 6, Issue 11
Abstract
Content Based Video retrieval is another facet of Content Based Retrieval systems. It is to be noted that videos are far more complex than images as they are a sequence of changing images and audio tracks. Here the content of the videos itself is matched with the query of the user. Video retrieval consists of a series of steps such as segmentation, frame selection, feature extraction and similarity measure. This paper is based on the fact that a video is a number of frames i.e. sequence of individual images which are moved at a certain speed. Therefore video is a sequence of moving images and same procedures used of Content based Image Retrieval for feature extraction can be applied to the video retrieval as well. The paper presents the approach called Color Difference Histogram which exploits the perceptually uniform color differences between two points under different background in accordance with the color and the edge orientations. As the system is based on the content of the videos, the query used here is a snap shot taken from the video recorder.
Authors and Affiliations
Anjali Sharma , Ajay Kumar Singh
Keywords
Related Articles
- EP ID EP164219
- DOI -
- Views 105
- Downloads 0
Modified 4-Bit Comparator Using Sleep Technique
The development of digital integrated circuits is challenged by higher power consumption. So low power circuits have become a top priority in modern VLSI design. This paper presents layout simulations of a new improved f...
LITERATURE REVIEW OF IMAGE COMPRESSION TECHNIQUE
Image compression is an efficient technique to reduce the size of graphical file and also reduce the storage requirement area. Image compression makes the faster transmission process as well as provides larger bandwidth...
Survey on a Novel approach for Reversible Data Hiding: Principles, Techniques and Recent studies
Today data hacking is a very big problem in networking field. There are many security techniques are available for solving the problem. One of these data-hiding, using this concept we can provide security, authentication...
A Novel Semi-Supervised Machine Learning Technique for Real-Time Network Traffic Classifications
Traffic classification technique is an essential tool for network and system security in the complex environments such as cloud computing based environment. The state-of-the-art traffic classification methods aim to take...
Closed Regular Pattern Mining Using Vertical Format
Discovering interesting patterns in transactional databases is often a challenging area by the length of patterns and number of transactions in data mining, which is prohibitively expensive in both time and space. Closed...