A Multi-Scale Image Detection Method for Power Line Inspection

DOI

10.2991/acsr.k.191223.027How to use a DOI?

Keywords

power line inspection, image processing, deep learning, data augmentation

Abstract

A multi-scale image detection method for power line inspection is provided by combining deep learning with traditional image algorithm. The method of Gauss blur and differential pyramid is used to enhance the data set. Experiments show that the method can effectively improve the efficiency of line inspection. Using FPGA to realize the Gauss blur and difference pyramid, using GPU to realize the Yolo V3, and using heterogeneous computing mode, a set of multi-scale image detection equipment is designed.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Nan Zhang
AU  - Fu Zhao
AU  - Jinglin Guo
PY  - 2019
DA  - 2019/12/24
TI  - A Multi-Scale Image Detection Method for Power Line Inspection
BT  - Proceedings of the 2019 International Conference on Big Data, Electronics and Communication Engineering (BDECE 2019)
PB  - Atlantis Press
SP  - 119
EP  - 121
SN  - 2352-538X
UR  - https://doi.org/10.2991/acsr.k.191223.027
DO  - 10.2991/acsr.k.191223.027
ID  - Zhang2019
ER  -