×
Home Current Archive Editorial board
Instructions for papers
For Authors Aim & Scope Contact
Original scientific article

ENHANCED INDUSTRIAL VISION SYSTEM BASED TEXTURE CLASSIFICATION USING DCTP AND DPNN MODELS

By
N.P. Rajeswari Orcid logo ,
N.P. Rajeswari

Assistant Professor, Department of Computer Science and Engineering, Velammal College of Engineering and Technology , Madurai, Tamil Nadu , India

M. Vijay Orcid logo
M. Vijay

Associate Professor, Department of Computer Science and Engineering, Kalasalingam Academy of Research and Education , Virudhunagar, Tamil Nadu , India

Abstract

The paper will discuss the issues with industrial vision systems in detecting defects, especially in the manufacturing of tyres, where changes in exterior light and distortion will compromise the accuracy of defect detection. The suggested scheme combines Directional Convoluted Ternary Pattern (DCTP) in the extraction of texture features and Distributed Pattern-based Neural Network (DPNN) in the classification. The DCTP model aims to complement the border detection of tyre defects, resolving the problem of illumination and noise variations. The DPNN classifier with a distributed method of learning patterns of features has a significant effect on the model complexity; the prediction accuracy is increased by fewer training samples. The model was statistically verified using sensitivity (0.9862), specificity (0.9743), precision (0.9801), recall (0.9862), F1-score (0.982), accuracy (98.28), and Kappa coefficient (0.972). The mentioned metrics demonstrate a substantial improvement over the current models, such as YOLOv5, CNN, and SVM. The suggested DCTP-DPNN architecture is better than the conventional methods with the highest classification accuracy (up to 98 %) and the lowest error (2 %). These results demonstrate that the hybrid feature extraction method on the basis of texture and classification using neural networks is good for detecting tyre defects in industries in real-time.

References

1.
Kazmi W, Nabney I, Vogiatzis G, Rose P, Codd A. An efficient industrial system for vehicle tyre (tire) detection and text recognition using deep learning. IEEE Transactions on Intelligent Transportation Systems. 2020 Jan 24;22(2):1264-75.
2.
Mohan P, Pahinkar A, Karajgi A, Kumar LD, Kasera R, Gupta AK, Narayanan SJ. Multi-contrast convolution neural network and fast feature embedding for multi-class tyre defect detection. In2020 4th International Conference on Electronics, Communication and Aerospace Technology (ICECA) 2020 Nov 5 (pp. 1397-1405). IEEE.
3.
Massaro A, Dipierro G, Cannella E, Galiano AM. Comparative analysis among discrete fourier transform, K-means and artificial neural networks image processing techniques oriented on quality control of assembled tires. Information. 2020 May 8;11(5):257.
4.
Harshitha A, Samala S. Detection of Tyre Wear Out Using OpenCV and Convolution Neural Networks– Survey. InCommunication Software and Networks: Proceedings of INDIA 2019 2020 Oct 4 (pp. 9-17). Singapore: Springer Singapore.
5.
Zhu J, Han K, Wang S. Automobile tire life prediction based on image processing and machine learning technology. Advances in Mechanical Engineering. 2021 Mar;13(3):16878140211002727.

Citation

This is an open access article distributed under the  Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 

Article metrics

Google scholar: See link

The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.