White whole (WW) grades cashew kernel’s classification using artificial neural network (ANN)

Narendra, V G and Shetty, Dasharathraj K (2018) White whole (WW) grades cashew kernel’s classification using artificial neural network (ANN). International Journal of Engineering and Technology(UAE), 7 (4). pp. 3442-3446. ISSN 2227-524X

[img] PDF
5427.pdf - Published Version
Restricted to Registered users only

Download (483kB) | Request a copy

Abstract

In this paper, we introduce an algorithm for the fitting of bounding rectangle to a closed region of cashew kernel in a given image. We propose an algorithm to automatically compute the coordinates of the vertices closed form solution. Which is based on coordinate geom-etry and uses the boundary points of regions. The algorithm also computes directions of major and minor axis using least-square ap-proach to compute the orientation of the given cashew kernel. More promising results were obtained by extracting shape features of a cashew kernel, it is proved that these features may predominantly use to make the better distinction of cashew kernels of different grades. The intelligent model was designed using Artificial Neural Network (ANN). The model was trained and tested using Back-Propagation learning algorithm and obtained classification accuracy of 89.74%.

Item Type: Article
Uncontrolled Keywords: Cashew Kernel; Bounding Rectangle; Shape Features; Artificial Neural Network.
Subjects: Engineering > MIT Manipal > Computer Science and Engineering
Depositing User: MIT Library
Date Deposited: 30 Nov 2018 05:46
Last Modified: 05 Apr 2019 10:49
URI: http://eprints.manipal.edu/id/eprint/152374

Actions (login required)

View Item View Item