请留言
slide1

Global Vision Media Focus

The company was featured on CCTV a total of 15 times, and its reach through Party media, central media, and local official media reached a total of 50 million people.

0101
News Categories

Satellite remote sensing extraction of burned areas

2024-10-28

Fire is a crucial natural disturbance factor in ecosystems, playing a vital role in maintaining ecosystem stability, promoting forest development, and controlling vegetation succession. Burned area is a key factor in fire assessment. The size of the burned area reflects the extent of the fire's impact on forest vegetation.

123 (2).png

Forest fire

Satellite monitoring has advantages such as large coverage area and high timeliness. It can conduct macro-monitoring of forest fire areas, provide means to monitor forest fire burning status, post-disaster losses and vegetation recovery, and thus provide scientific and technological support for decision-making work such as post-disaster recovery and reconstruction.

123 (3).png

(a) Before the disaster

123 (4).png

 (b) Remote sensing monitoring map of the burned area after the disaster

This algorithm uses two images covering the same area before and after a fire to calculate the burned area based on the Global Vegetation Index Monitoring (GEMI_B) algorithm. Because vegetation regrowth after a fire can prevent accurate extraction of the fire-affected area, the algorithm requires that the selected post-fire image be as close as possible to the time of the fire.

The GEMI_B index typically shows significant differences before and after a fire, a property that can be used to extract the burned area and calculate the burned area. Because the GEMI_B index effectively distinguishes burned areas from vegetation, clouds, and water bodies, it preserves the characteristics of burned areas and avoids the weakening of burned area information due to vegetation recovery.

The formula for calculating GEMI_B is:

123 (6).png

123 (7).png

The GEMI_B method is used to monitor the burned area, with the following criteria:

GEMI_B1-GEMI_B2> 0 & ρ1

GEMI_B1 and GEMI_B2 are the GEMI_B values ​​before and after the fire, respectively.

After identifying the burned pixels, the image needs to undergo an ecological closing operation to fill in small break points and cloud cover points, followed by an ecological opening operation to remove isolated false burned areas, and finally the burned area is calculated.

The fire-affected area obtained by inversion based on this method is shown in the figure below.

123 (5).jpg

 Results extracted from the burned area