Redirigiendo al acceso original de articulo en 15 segundos...
ARTÍCULO
TITULO

Residual Multi-Attention Classification Network for A Forest Dominated Tropical Landscape Using High-Resolution Remote Sensing Imagery

Tong Yu    
Wenjin Wu    
Chen Gong and Xinwu Li    

Resumen

Tropical forests are of vital importance for maintaining biodiversity, regulating climate and material cycles while facing deforestation, agricultural reclamation, and managing various pressures. Remote sensing (RS) can support effective monitoring and mapping approaches for tropical forests, and to facilitate this we propose a deep neural network with an encoder?decoder architecture here to classify tropical forests and their environment. To deal with the complexity of tropical landscapes, this method utilizes a multi-scale convolution neural network (CNN) to expand the receptive field and extract multi-scale features. The model refines the features with several attention modules and fuses them through an upsampling module. A two-stage training strategy is proposed to alleviate misclassifications caused by sample imbalances. A joint loss function based on cross-entropy loss and the generalized Dice loss is applied in the first stage, and the second stage used the focal loss to fine-tune the weights. As a case study, we use Hainan tropical reserves to test the performance of this model. Compared with four state-of-the-art (SOTA) semantic segmentation networks, our network achieves the best performance with two Hainan datasets (mean intersection over union (MIoU) percentages of 85.78% and 82.85%). We also apply the new model to classify a public true color dataset which has 17 semantic classes and obtain results with an 83.75% MIoU. This further demonstrates the applicability and potential of this model in complex classification tasks.

 Artículos similares

       
 
Julio Cesar Pascale Palhares, Danielle Leal Matarim, Rafael Vieira de Sousa and Luciane Silva Martello    
The aim of the study is to discern benchmarks for the indicators L water cow-1 day-1 and L water kg milk-1 day-1 per type of production system and season. A total of 876 commercial dairy farms underwent comprehensive water consumption monitoring from Jan... ver más
Revista: Water

 
Shichao Wang, Jun Song, Junru Guo, Yanzhao Fu, Yu Cai and Linhui Wang    
As one of the most significant disturbance sources in the upper marine environment of the South China Sea, tropical cyclones (typhoons) serve as a typical research subject for investigating the energy transfer process between the ocean and atmosphere. Ut... ver más
Revista: Water

 
Tomás R. Bolaño-Ortiz, Viverlys L. Díaz-Gutiérrez, Andrés M. Vélez-Pereira, Eliana L. Vergara-Vásquez and Yiniva Camargo-Caicedo    
This article investigates the snow albedo changes in Colombian tropical glaciers, namely, Sierra Nevada de Santa Marta (SNSM), Sierra Nevada del Cocuy (NSC), Nevado del Ruíz (NDR), Nevado Santa Isabel (NDS), Nevado del Tolima (NDT), and Nevado del Huila ... ver más
Revista: Water

 
Nichapa Parasin and Teerachai Amnuaylojaroen    
A heat index is a key indicator directly related to meteorological factors influencing human health, particularly work performance. However, the interaction between air quality, meteorology, heat, and associated work performance is loosely defined, espec... ver más
Revista: Urban Science

 
Tomeu Rigo    
The Metropolitan Area of Barcelona is one of the most densely populated European regions. The hydrological conditions are very extreme, as are those of the Mediterranean Basin, with long drought periods. The management of water systems is one of the prio... ver más
Revista: Hydrology