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

Automated Object Detection and Sorting

Author(s):

Aashay Bane ; Avinash Panicker; Nandan Naik; Rakesh Malviya; Dr.A.M.Salsingikar

Keywords:

Automated Object Detection, Object Sorting

Abstract

In many industries object sorting techniques uses maximum advantages of imaging technologies such as hyper spectral technique. In today’s world, application of image processing in many industrial purposes has proven its acceptance and supremacy. Because of multi-level nature design of object sorting algorithms is a challenging pattern recognition problem. Objects represented by sets of pixels per spectra in hyper spectral images are to be allocated into pre-specified sorting categories. This Project illustrates the designing of two-stage algorithms of sorting, learning to distinguish individual pixels per spectra and combining the decisions of per pixel into a single per object outcome using MATLAB software. The Project provides a case-study on designs of algorithms in a real-world industrial sorting problem using webcam. Depending upon the previous knowledge of image sorting techniques, four algorithms are studied. Assuming the ideal system, the sorting accuracy as well as the algorithm execution speed is estimated. We discuss the accuracy or speed of different algorithms.

Other Details

Paper ID: NCTAAP156
Published in: Conference 4 : NCTAA 2016
Publication Date: 00/00/0000
Page(s): 673-676

Article Preview




Download Article