Forest Fire Smoke Detection Using Back-Propagation Neural Network Based on MODIS Data

Li, Xiaolian; Song, Weiguo; Lian, Liping; Wei, Xiaoyong

doi:10.3390/rs70404473

Cited by 63 publications

(63 citation statements)

References 55 publications

(81 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The data of the MODIS sensor mounted on Terra and Aqua satellites have been widely used for wildfire detection, monitoring, and assessment [3,6,13,[50][51][52]. Since the launch of the Terra satellite in December 1999 [53] and the Aqua satellite in May 2002 [54], a large amount of MODIS data of wildfires has been captured during the past two decades.…”

Section: Data Sourcementioning

confidence: 99%

“…Cloud: Cloud is the most common class in satellite imageries. It has the similar shape, color, and spectral features in comparison to smoke [6]. Meanwhile, smoke frequently mixes with cloud when wildfires occur.…”

Section: Classesmentioning

confidence: 99%

“…With the rapid development in the past decades, satellite remote sensing brings a great opportunity for this task with the advantages of timeliness, wide observation, and low cost [5]. Nevertheless, the identification of smoke using satellite data is challenging because the fire smoke has varying shapes, colors, scopes, and spectral overlaps [2,3,6]. This makes it difficult data, namely USTC_SmokeRS, by considering smoke-related classes.…”

Section: Introductionmentioning

confidence: 99%

“…[3] investigated different detecting schemes for the smoke above the land and ocean using spectral and spatial threshold sets. Despite that the fixed thresholds for multiple bands may be valid in local regions, it is difficult to determine the optimal thresholds because of the spatial and temporal variations [6]. This can lead to the small regions/areas of smoke being neglected, which reduces the timeliness of fire alarm.…”

mentioning

confidence: 99%

“…This can lead to the small regions/areas of smoke being neglected, which reduces the timeliness of fire alarm. In addition, Li et al [14] developed an automatic algorithm to detect smoke area by K-means clustering and fisher linear discrimination, and Li et al [2,6] explored the neural network to identify smoke pixels in the image. These methods used the training samples of a few classes besides smoke, such as cloud, water, and vegetation.…”

mentioning

confidence: 99%

See 4 more Smart Citations

SmokeNet: Satellite Smoke Scene Detection Using Convolutional Neural Network with Spatial and Channel-Wise Attention

et al. 2019

Self Cite

View full text Add to dashboard Cite

A variety of environmental analysis applications have been advanced by the use of satellite remote sensing. Smoke detection based on satellite imagery is imperative for wildfire detection and monitoring. However, the commonly used smoke detection methods mainly focus on smoke discrimination from a few specific classes, which reduces their applicability in different regions of various classes. To this end, in this paper, we present a new large-scale satellite imagery smoke detection benchmark based on Moderate Resolution Imaging Spectroradiometer (MODIS) data, namely USTC_SmokeRS, consisting of 6225 satellite images from six classes (i.e., cloud, dust, haze, land, seaside, and smoke) and covering various areas/regions over the world. To build a baseline for smoke detection in satellite imagery, we evaluate several state-of-the-art deep learning-based image classification models. Moreover, we propose a new convolution neural network (CNN) model, SmokeNet, which incorporates spatial and channel-wise attention in CNN to enhance feature representation for scene classification. The experimental results of our method using different proportions (16%, 32%, 48%, and 64%) of training images reveal that our model outperforms other approaches with higher accuracy and Kappa coefficient. Specifically, the proposed SmokeNet model trained with 64% training images achieves the best accuracy of 92.75% and Kappa coefficient of 0.9130. The model trained with 16% training images can also improve the classification accuracy and Kappa coefficient by at least 4.99% and 0.06, respectively, over the state-of-the-art models.

show abstract