High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Extended LCCF for Object Detection

Author(s):

Disha Davis , IES College of Engineering, Chittilappilly, Thrissur; Nicy Johnson, IES College of Engineering, Chittilappilly, Thrissur

Keywords:

Object Detection, Correlation Filters, SADMM, Extended LCCF, Autopath, Feature Extraction, Classifiers

Abstract

Object detection is a series of segments containing the features of interest, which are taken as pre-processing and widely applied in various vision tasks. Of these, face recognition is one of its major application. It is a computer vision technology related to the modern method called image processing that deals with identifying instances in the objects of a certain class such as humans that get resided in both images and videos. However, most of existing approaches only utilizes the proposals to compute the location; the aim is to propose a fastest algorithm for making the process as an efficient task. So the correlation procedure has been introduced in order to overcome these existing disadvantages. It is a highly accurate method in order to solve the optimization problems in the correlation filters, more specifically, filters called extended latent constrained correlation filters (LCCF) are proposed. Extended LCCF will maps the correlation filters to a given latent subspace that is created thereby establish a new learning framework. It will relate each pixel in the real time video that is given as the input and will extract the feature points by feature extraction method. The feature points will be stored in the subspace where it recreates using a newer algorithm called autopath algorithm. It will accurately detect the person in the video when it contains noises and occlusions. The extended LCCF method will outperform significantly better than other competing approaches.

Other Details

Paper ID: IJSRDV7I40705
Published in: Volume : 7, Issue : 4
Publication Date: 01/07/2019
Page(s): 725-728

Article Preview

Download Article