ARTÍCULO
TITULO

DFFAN: Dual Function Feature Aggregation Network for Semantic Segmentation of Land Cover

Junqing Huang    
Liguo Weng    
Bingyu Chen and Min Xia    

Resumen

Analyzing land cover using remote sensing images has broad prospects, the precise segmentation of land cover is the key to the application of this technology. Nowadays, the Convolution Neural Network (CNN) is widely used in many image semantic segmentation tasks. However, existing CNN models often exhibit poor generalization ability and low segmentation accuracy when dealing with land cover segmentation tasks. To solve this problem, this paper proposes Dual Function Feature Aggregation Network (DFFAN). This method combines image context information, gathers image spatial information, and extracts and fuses features. DFFAN uses residual neural networks as backbone to obtain different dimensional feature information of remote sensing images through multiple downsamplings. This work designs Affinity Matrix Module (AMM) to obtain the context of each feature map and proposes Boundary Feature Fusion Module (BFF) to fuse the context information and spatial information of an image to determine the location distribution of each image?s category. Compared with existing methods, the proposed method is significantly improved in accuracy. Its mean intersection over union (MIoU) on the LandCover dataset reaches 84.81%.

 Artículos similares

       
 
Wenxin Qiao, Hao Lu, Yu Lu, Lijie Meng and Yicen Liu    
Satellite?terrestrial integrated networks (STINs) are regarded as a promising solution to meeting the demands of global high-speed seamless network access in the future. Software-defined networking and network function virtualization (SDN/NFV) are two co... ver más
Revista: Future Internet

 
Dayal Wijayarathne, Paulin Coulibaly, Sudesh Boodoo and David Sills    
Demand for radar Quantitative Precipitation Estimates (QPEs) as precipitation forcing to hydrological models in operational flood forecasting has increased in the recent past. It is practically impossible to get error-free QPEs due to the intrinsic limit... ver más
Revista: Water

 
Xin Song, Yue Ni, Xiuwei Han, Lei Qin and Li Dong    
In this paper, we propose an optimal time and power allocation scheme in a wireless power supply full-duplex (FD) relay system, where we consider the number of relay antennas in the energy harvesting stage. At the same time, the energy efficiency optimiz... ver más
Revista: Future Internet

 
Martin N. Hjelmeland, Arild Helseth and Magnus Korpås    
We propose a model for medium-term hydropower scheduling (MTHS) with variable head and uncertainty in inflow, reserve capacity, and energy price. With an increase of intermittent energy sources in the generation mix, it is expected that a flexible hydrop... ver más
Revista: Energies

 
This paper proposes a harmonic compensation control with disturbance rejection function for a standalone inverter. Due to the LC type three-phase three-leg inverter is connected to nonlinear loads, low-order harmonic components appears in the inverter ou... ver más
Revista: Energies