Abstract
The efficient and accurate application of deep learning in the remote sensing field largely depends on the pre-processing technology of remote sensing images. Particularly, image fusion is the essential way to achieve the complementarity of the panchromatic band and multispectral bands in high spatial resolution remote sensing images. In this paper, we not only pay attention to the visual effect of fused images, but also focus on the subsequent application effectiveness of information extraction and feature recognition based on fused images. Based on the WorldView-3 images of Tongzhou District of Beijing, we apply the fusion results to conduct the experiments of object recognition of typical urban features based on deep learning. Furthermore, we perform a quantitative analysis for the existing pixel-based mainstream fusion methods of IHS (Intensity-Hue Saturation), PCS (Principal Component Substitution), GS (Gram Schmidt), ELS (Ehlers), HPF (High-Pass Filtering), and HCS (Hyper spherical Color Space) from the perspectives of spectrum, geometric features, and recognition accuracy. The results show that there are apparent differences in visual effect and quantitative index among different fusion methods, and the PCS fusion method has the most satisfying comprehensive effectiveness in the object recognition of land cover (features) based on deep learning.
Highlights
With the development of earth observation technology, a large number of remote sensing satellites have been launched, which further improves the acquisition ability of high spatial resolution and high spectral resolution imagery, and provides extensive data sources for applications [1]
Object recognition of urban typical land features from High Spatial Resolution Remote Sensing Imagery (HSRRSI) is an active and important research task driven by many practical applications
High spatial resolution remotely sensed imagery often contains multiple types of land-cover with distinct spatial, spectral, and geometric characteristics, and the manual labeling sample is not enough, which limits the applications of deep learning in object recognition from HSRRSI [6]
Summary
With the development of earth observation technology, a large number of remote sensing satellites have been launched, which further improves the acquisition ability of high spatial resolution and high spectral resolution imagery, and provides extensive data sources for applications [1]. High spatial resolution remotely sensed imagery often contains multiple types of land-cover with distinct spatial, spectral, and geometric characteristics, and the manual labeling sample is not enough, which limits the applications of deep learning in object recognition from HSRRSI [6]. Six traditional spatial–spectral fusion methods are selected for panchromatic and multispectral bands in the study area to generate remote sensing images with both high spatial resolution and high spectral resolution. It verifies that a fusion image with high spatial resolution and high spectral resolution achieves better recognition effect
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
