Man-made target detection in the natural background is the core problem of high-resolution optical remote sensing image interpretation, and it is an important application direction in the target recognition, situation awareness, and other fields. This paper proposes a man-made target detection method based on red-edge spectral information in the natural background. In this method, the red-edge spectrum information of vegetation is used to suppress background interference and narrow the search area in large remote sensing images, to improve the probability of target detection. The automatic extraction model of man-made targets is trained by the public dataset COD10K and self-constructed remote sensing dataset, and detection experiments are carried out by using the multispectral images from the WorldView-3 satellite and the Sentinel-2 A/B satellites. The experimental results show that the proposed algorithm has high detection accuracy and fast detection speed for the green runway, green buildings, and another man-made targets in the forest or grassland. Compared with the traditional man-made target detection algorithm, this method is more suitable for the rapid search of large area remote sensing images.
The high-resolution multi-mode imaging satellite called GFDM (Gao Fen Duo Mo) has been successful launched in July 3 of 2020, which was integrated with 1 panchromatic and 8 multispectral bands with the spatial resolution of 0.42 m and 1.6 m, respectively. The Synchronization Monitoring Atmospheric Corrector (SMAC) instrument was also on board GFDM satellite, aiming for offering time synchronized and field of view overlapped atmospheric measurements to improve the atmospheric correction of the GFDM main sensor image. As the first civilian atmospheric corrector onboard high spatial resolution satellite with polarization detection, SMAC has 8 wavelength bands from visible to short-wave infrared with the spatial resolution about 6.7 km. In order take full advantage of the multispectral measurements of SMAC, we investigate to retrieve the aerosol optical depth (AOD) by using the intensity measurements in this study. To decouple the surface-atmosphere contribution from SAMC measurements, the corresponding surface reflectance over land is derived from the Moderate-resolution Imaging Spectroradiometer (MODIS) surface bi-directional reflectance climatology. Based on the principal component analysis method and the dataset from spectral libraries, the surface reflectance ratios are further obtained by spectral conversion with the spectral response function from MODIS to SAMC for aerosol retrieval. With the aerosol look-up table (LUT) established by the Second Simulation of the Satellite Signal in the Solar Spectrum (6S) radiation transmission model, the multispectral inversions are carried out and the AODs are retrieved. In addition, the AOD data from Aerosol Robotic Network are used to validate the retrieved results from SAMC by the spatial-temporal matching, the statistical parameters including the root mean square error (RMSE) and the correlation coefficient (R) are employed together. By this means, the retrieved AODs from the intensity measurements of SAMC are preliminary investigated.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.