Abstract

The Visible Infrared Imaging Radiometer Suite (VIIRS) Day/Night Band (DNB) is a panchromatic visible and near-infrared band capable of quantitative measurement of light radiances from 3×10-9 W∙cm-2∙sr-1 to 2×10-2 W∙cm-2∙sr-1. Its extreme sensitivity of the HGS to low lights, together with the nearly constant spatial resolution over the whole swath, enables numerous applications of environmental remote sensing and global monitoring of anthropogenic activities in nighttime. However, the DNB is susceptible to high energy particles (HEPs) hitting over the South Atlantic Anomaly (SAA) region. Although a special algorithm was designed to filter out high radiance pixels induced by HEPs, large amount of artificial bright pixels are still observed in the generated DNB nighttime images, leaving obstacles for several image based applications, such as automatic fishing boat detection over the SAA region.

In this study, we investigated DNB nighttime images over the SAA region. It has been shown that the DNB gain selection logic is the possible root cause for the isolated bright pixels in DNB nighttime images over the SAA region. This result was confirmed by a recent DNB special data acquisition over the SAA region. With the knowledge acquired from the analysis of root cause, we explored method to eliminate artificial bright pixels induced by the HEPs over the SAA region. Our method could improve the DNB nighttime image quality over the SAA region and potentially benefits monitoring of anthropogenic activities in nighttime over that region.

Share

COinS
 
Aug 24th, 9:50 AM

Elimination of Artificial Bright Pixels in VIIRS DNB Nighttime IMage over the South Atlantic Anomaly Region

The Visible Infrared Imaging Radiometer Suite (VIIRS) Day/Night Band (DNB) is a panchromatic visible and near-infrared band capable of quantitative measurement of light radiances from 3×10-9 W∙cm-2∙sr-1 to 2×10-2 W∙cm-2∙sr-1. Its extreme sensitivity of the HGS to low lights, together with the nearly constant spatial resolution over the whole swath, enables numerous applications of environmental remote sensing and global monitoring of anthropogenic activities in nighttime. However, the DNB is susceptible to high energy particles (HEPs) hitting over the South Atlantic Anomaly (SAA) region. Although a special algorithm was designed to filter out high radiance pixels induced by HEPs, large amount of artificial bright pixels are still observed in the generated DNB nighttime images, leaving obstacles for several image based applications, such as automatic fishing boat detection over the SAA region.

In this study, we investigated DNB nighttime images over the SAA region. It has been shown that the DNB gain selection logic is the possible root cause for the isolated bright pixels in DNB nighttime images over the SAA region. This result was confirmed by a recent DNB special data acquisition over the SAA region. With the knowledge acquired from the analysis of root cause, we explored method to eliminate artificial bright pixels induced by the HEPs over the SAA region. Our method could improve the DNB nighttime image quality over the SAA region and potentially benefits monitoring of anthropogenic activities in nighttime over that region.