Table of Content

Open Access iconOpen Access

ARTICLE

crossmark

Multi-Scale Boxes Loss for Object Detection in Smart Energy

Zhiyong Dai1,*, Jianjun Yi1, Yajun Zhang1, Liang He2

1 School of Mechanical and Power Engineering, East China University of Science and Technology, Shanghai, 200237, China
2 Shanghai Aerospace Control Technology Institute, Shanghai, 201109, China

* Corresponding Author: Zhiyong Dai. Email: email

Intelligent Automation & Soft Computing 2020, 26(5), 887-903. https://doi.org/10.32604/iasc.2020.010122

Abstract

The rapid development of Internet of Things (IoT) technologies has boosted smart energy networks in recent years. However, power line surveillance systems still suffer from the low accuracy and efficiency of the power line area recognition and risk objects detection. This paper proposes a new customized loss function to tackle the disequilibrium of the size of objects on multi-scale feature maps in the deep learning-based detectors. To validate the new concept and improve the efficiency, we also presented a new object detection model. Experimental results are provided to exhibit the advantage of our proposed method in both accuracy and efficiency.

Keywords


Cite This Article

APA Style
Dai, Z., Yi, J., Zhang, Y., He, L. (2020). Multi-scale boxes loss for object detection in smart energy. Intelligent Automation & Soft Computing, 26(5), 887-903. https://doi.org/10.32604/iasc.2020.010122
Vancouver Style
Dai Z, Yi J, Zhang Y, He L. Multi-scale boxes loss for object detection in smart energy. Intell Automat Soft Comput . 2020;26(5):887-903 https://doi.org/10.32604/iasc.2020.010122
IEEE Style
Z. Dai, J. Yi, Y. Zhang, and L. He, “Multi-Scale Boxes Loss for Object Detection in Smart Energy,” Intell. Automat. Soft Comput. , vol. 26, no. 5, pp. 887-903, 2020. https://doi.org/10.32604/iasc.2020.010122

Citations




cc Copyright © 2020 The Author(s). Published by Tech Science Press.
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
  • 1652

    View

  • 1135

    Download

  • 1

    Like

Share Link