A deep neural network supported by ground verification data for active fire detection in MODIS

Jinpeng Chen; Feifei Xie; Chunkai Zheng

doi:10.1117/12.2681300

8 June 2023 A deep neural network supported by ground verification data for active fire detection in MODIS

Jinpeng Chen, Feifei Xie, Chunkai Zheng

Proceedings Volume 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023); 127074A (2023) https://doi.org/10.1117/12.2681300
Event: International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 2023, Changsha, China

Abstract

At present, the use of MODIS to detect fires mainly uses a series of thresholds to identify fire. However, in areas with high heterogeneity, the division of thresholds is more difficult, resulting in many false detections of fire, and cold fire are easily missed. The emergence of deep learning technology has improved the classification accuracy of images, but the surface information of hotspots cannot be determined only by visual interpretation of MODIS images, and the acquisition of highquality fire samples further limits the detection performance of deep learning methods. In order to solve the above problems, this paper proposes a deep neural network fire detection method based on ground truth data. First, the data structure of the US ground wildfire dataset is studied, a spatiotemporal matching strategy between ground data and remote sensing images is designed, and a priori sample is constructed. library. Second, robust classification features are selected combining the fire physical properties and the spectral distribution of the dataset. Finally, a DNN (Deep Neural Network) for MODIS fire detection is constructed according to the image characteristics. Application experiments were carried out in three typical scenarios and compared with MODIS fire products. The results show that the improved method improves the temporal and spatial adaptability of fire detection, and the extraction of small fire in farming areas is better than MODIS fire products; especially in the suburbs Significant reduction of spurious fire in region extraction,false positives decreased by 19.8%. This method provides a new reference for coarse resolution satellite data to detect fires, and effectively improves the efficiency of fire detection.

Citation Download Citation

Jinpeng Chen, Feifei Xie, and Chunkai Zheng "A deep neural network supported by ground verification data for active fire detection in MODIS", Proc. SPIE 12707, International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023), 127074A (8 June 2023); https://doi.org/10.1117/12.2681300

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Fire

Forest fires

MODIS

Detection and tracking algorithms

Satellites

Deep learning

Neural networks

Show All Keywords

Keywords/Phrases

Search In:

Publication Years