ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Real Time Object Detection and 3D Modeling Using Fuzzy Logic

Journal: International Journal of Science and Research (IJSR) (Vol.3, No. 6)

Publication Date:

Authors : ; ;

Page : 1173-1177

Keywords : OD3DM(Object Detection & 3D Modeling); Entropy; fuzzy Logic; wavelets; Centroid based Similarity Matrix; pdf (probability distribution functions);

Source : Downloadexternal Find it from : Google Scholarexternal


This paper OD3DM (Object detection and 3D modeling) mainly discussed the process to detect complex geometric objects and thereafter performing 3D modeling of geometric objects using Entropy based selection of optimum transformation of input data, wavelet based transformation and fuzzy logictechniques for designing and training of object recognition systems using realistic 3D computer graphics models using Fuzzy Logic has been implemented in this paper. Our model OD3DM system convert picture into grid of size 10x10.10x20, 20x20 and uses centroid formula to detect geometric objects which detect edge and boundary limit of every image taken from real time camera.We look at the relation between the size of the training set and the classification accuracy for a basic recognition task and provide a method for estimating the degree of difficulty of detecting an object. A few images were taken and were captured using real time camera. We show how to sample, align, detect, and rotate images of objects. We address the problem of training on large, highly redundant data and propose a novel active learning method which generates compact training sets and compact classifiers. We believe that the use of realistic 3D models and computer graphics for view-based object recognition will lead to a reevaluation of some of the basic research questions in this area. The problems of learning from small training sets and learning from data with inaccurate or missing ground truth will lose importance while the problem of learning from large, accurately labeled but highly redundant data will become more important.

Last modified: 2014-06-25 20:59:17