slugA Semantic Concept Detection for Video Based on Regularized Extreme Learning Machine
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2014, Vol 2, Issue 4
Abstract
This paper explains the concept detection based on the proposed method regularized extreme learning machine. In regularized extreme learning machine deals with the modification of the extreme learning machine to solve the missing data problems. Semantic concept detection is an important step in concept- based semantic video retrieval, which can be regarded as an intermediate descriptor to bridge the semantic gap. Support Vector Machines (SVM) and ELM (Extreme Learning Machine) is most existing methods. However, there are several drawbacks of using SVM, such as the high computational cost and large number of parameters to be optimized. The drawback facing by using ELM is some parameters are needed to be tuned manually. This consumes time for classification process. Instead of these disadvantages we use a proposed ELM called Regularized extreme learning machine (RELM) is used to detect semantic concept of videos. It uses a cascade of L1 penalty (LARS) and L2 penalty (Tikhonov regularization) on ELM (TROP-ELM) to regularize the matrix computation.
Authors and Affiliations
Julin Rose Jacob
Keywords
Related Articles
- EP ID EP17832
- DOI -
- Views 340
- Downloads 12
EEG Signal Classification and Drowsiness Detection in Driver
traffic accidents are increasing due to driver’s consumption of alcohol or due to drowsiness. Breathalyzers are used for the purpose of alcohol detection, these device measure the amount of alcohol rather than loss of f...
Power Quality Enhancement in Grid connected Microgrid While Supplying Non-Linear Loads
The objective of the project shows how the power quality can be improved in a microgrid that is supplying a nonlinear load. The microgrid contains a hybrid combination of inertial and converter interfaced distributed ge...
Development of an Inbuilt UHF Signal and Weather Parameters Measuring System
This paper presents the development of an inbuilt measuring system suitable for measuring UHF signal strength and weather parameters concurrently for accurate prediction of the effects of weather parameters on UHF radio...
Survey Paper on Circular Aperture Slot Antenna with Defected Ground Structure for Broad Band
This paper introduces the survey on the circular Aperture Sot Antenna with defected ground structure. A novel system composed of a circular aperture slot antenna and a Common-Mode (CM) noise rejection filter is presente...
Study of Mechanical Characteristics on Hybrid Composites Using Sisal Fiber and Banana Fiber
Fiber reinforced polymer composites have a wide variety of applications as a class of structural materials because of their advantages such as ease of manufacturing, comparatively minimum cost of production & greater st...