A novel approach to machine learning for object detection and recognition

DOI

10.2991/978-94-6463-471-6_87How to use a DOI?

Keywords

RCNN algorithm; YOLO; Optimization; Neural Networks; Computer Vision; Image Processing

Abstract

Artificial intelligence (AI) in computer science is a branch that focuses on creating intelligent systems or robots capable of imitating human behavior and reactions. Artificial intelligence is a branch of computer science. The impressive ability of individuals to easily identify and differentiate items using their visual senses is astonishing. However, robots have many substantial obstacles in detecting and identifying objects. Neural networks are a suggested solution from the field of computer science to address this problem. Additionally, it is sometimes known as “Artificial Neural Networks.“ These words are all used interchangeably. Neural networks are an example of artificial intelligence that operates without the usage of symbols. The purpose of these computer models is to simulate the operations of the human brain to assist in the identification and classification of various objects. Conversely, object detection and identification is a field that undergoes extensive investigation. This research focuses mostly on dynamic things, which are objects in motion. The system is meant to perform object detection and static object identification to achieve its specified purposes. Our solution replaces the basic classifier from the previous system with a more advanced one, resulting in an increased accuracy rate. The Object Detection and Tracking System will use the renowned deep learning network Faster Regional Convolution Neural Network (Faster R-CNN) for object detection. This implies that the system will have the capability to identify items. Moreover, the traditional object tracking mechanism will be used in this specific project. By using closed-circuit television cameras in tunnels, it will enable the automated detection and recording of any unforeseen events. The Object Detection and Tracking System used deep learning to carry out its functions. This model was trained on a dataset consisting of tunnel event photos captured by photography. The model attained mean accuracy values of 0.8479, 0.7161, and 0.9085 for detecting fire target items. The values were acquired for detecting autos, persons, and firing targets. The model generated these numbers via its computations. The Tunnel CCTV Accident Detection System, based on ODTS, was tested by analyzing four accident videos, each containing a separate accident, using a trained deep learning model.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Bh. Sai Venkata Ganesh
AU  - N. Siva Kumar
PY  - 2024
DA  - 2024/07/30
TI  - A novel approach to machine learning for object detection and recognition
BT  - Proceedings of the International Conference on Computational Innovations and Emerging Trends (ICCIET- 2024)
PB  - Atlantis Press
SP  - 918
EP  - 925
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-471-6_87
DO  - 10.2991/978-94-6463-471-6_87
ID  - Ganesh2024
ER  -