Searching Damaged Pine Trees by Wilt Disease Based on Deep Learning Using Multispectral Image 


Vol. 45,  No. 11, pp. 1823-1830, Nov.  2020
10.7840/kics.2020.45.11.1823


PDF
  Abstract

Wilt disease is a cause of massive damage to pine forests in east Asia including Korea, Japan, and China. One of the effective ways to prevent pine trees from the spread of wilt disease is early detection and removal of damaged trees. This paper proposes a searching method of the damaged pine trees by wilt disease in ortho-images corrected from 5-band multi-spectral images, which are captured by unmanned aviation vehicles. The proposed method constructs patch-based classifier because of the lack of data using ResNet18 backbone network, classifies the 5-band multi-spectral(RGB, NIR, RedEdge) ortho-image patches, and make the results as a heat map. The resulted heat map can be used to find the approximate location and distribution of diseased pine trees to show the trend of spreading disease, and to extract the spectral distribution of the diseased areas in the image. The patch-based classifier with the multispectral images shows 97.06% of accuracy, which is little better than 96.50% with RGB images.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

ZHANGRUIRUI, YOUJIE, D. Kim, S. A. Lee, J. Lee, "Searching Damaged Pine Trees by Wilt Disease Based on Deep Learning Using Multispectral Image," The Journal of Korean Institute of Communications and Information Sciences, vol. 45, no. 11, pp. 1823-1830, 2020. DOI: 10.7840/kics.2020.45.11.1823.

[ACM Style]

ZHANGRUIRUI, YOUJIE, Dong-Hoon Kim, Sol Ae Lee, and Joonwhoan Lee. 2020. Searching Damaged Pine Trees by Wilt Disease Based on Deep Learning Using Multispectral Image. The Journal of Korean Institute of Communications and Information Sciences, 45, 11, (2020), 1823-1830. DOI: 10.7840/kics.2020.45.11.1823.

[KICS Style]

ZHANGRUIRUI, YOUJIE, Dong-Hoon Kim, Sol Ae Lee, Joonwhoan Lee, "Searching Damaged Pine Trees by Wilt Disease Based on Deep Learning Using Multispectral Image," The Journal of Korean Institute of Communications and Information Sciences, vol. 45, no. 11, pp. 1823-1830, 11. 2020. (https://doi.org/10.7840/kics.2020.45.11.1823)

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880