Open Access

ARTICLE

An Ontology Based Multilayer Perceptron for Object Detection

P. D. Sheena Smart^1,*, K. K. Thanammal², S. S. Sujatha²

1 Department of Computer Science, S. T. Hindu College, Nagercoil, Affiliated to Manonmaniam Sundaranar University, Abishekapatti, Tirunelveli, 627012, Tamilnadu, India
2 Department of Computer Science and Applications, S. T. Hindu College, Nagercoil, 629002, India

* Corresponding Author: P. D. Sheena Smart. Email:

Computer Systems Science and Engineering 2023, 44(3), 2065-2080. https://doi.org/10.32604/csse.2023.028053

Received 01 February 2022; Accepted 03 March 2022; Issue published 01 August 2022

Abstract

In object detection, spatial knowledge assisted systems are effective. Object detection is a main and challenging issue to analyze object-related information. Several existing object detection techniques were developed to consider the object detection problem as a classification problem to perform feature selection and classification. But these techniques still face, less computational efficiency and high time consumption. This paper resolves the above limitations using the Fuzzy Tversky index Ontology-based Multi-Layer Perception method which improves the accuracy of object detection with minimum time. The proposed method uses a multilayer for finding the similarity score. A fuzzy membership function is used to validate the score for predicting the burned and non-burned zone. Experimental assessment is performed with different factors such as classification rate, time complexity, error rate, space complexity, and precision by using the forest fire dataset. The results show that this novel technique can help to improve the classification rate and reduce the time and space complexity as well as error rate than the conventional methods.

---

Keywords

---