Abstract
Due to the limited training data, current data-driven algorithms, including deep convolutional networks (DCNs), are susceptible to training data that cannot be applied to new data directly. Unlike existing methods that are trying to improve model generation capability using limited data, we introduce a learning-based image translation method to generate data that share the same characteristics of target data. The low-resolution panchromatic satellite images are converted into high-resolution color images through interpolation and colorization with the proposed symmetric colorization network (SCN). Experiments on a very-high-resolution (VHR) dataset show that images generated by our SCN are with both quantitatively and qualitatively high color fidelity. Furthermore, we also demonstrate that high extraction accuracy is retained during the model transferring from aerial to satellite images. For pre-trained feature pyramid network (FPN), compared to the performance on raw panchromatic images, the interpolated and colorized images increase 305.7% of recall (0.929 vs. 0.229), 78.2% of overall accuracy (0.768 vs. 0.431), 132.5% of f1-score (0.851 vs. 0.366), and 230.8% of Jaccard index (0.741 vs. 0.224), respectively.
Highlights
Automatic extraction of various land-covers, such as buildings, roads, and trees, is a long-existing demand in the field of remote sensing
As for the unbalanced metrics, the method shows the highest values of precision and recall on panchromatic images and colorized images with our symmetric colorization network (SCN) model, respectively
Among all colorized satellite images, the image generated by our SCN shows the highest values of recall, overall accuracy, f1-score, and Jaccard index
Summary
Automatic extraction of various land-covers, such as buildings, roads, and trees, is a long-existing demand in the field of remote sensing. With the dramatic increase in the availability and accessibility of very high resolution (VHR) remote sensing imagery, efficient and accurate methods have become increasingly urgent [1]. Existing algorithms can be classified into two categories depending on whether they require training data or not: (1) unsupervised and (2) supervised methods [2]. Automatic extraction is performed by mathematical manipulation of image characteristics, such as thresholding of pixel values or histograms [3], edge detection [4], and region analysis [5]. While sufficient learning data are available, The associate editor coordinating the review of this manuscript and approving it for publication was Wenming Cao
Sign-in/Register to view highlights & summary 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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
