Technology‎ > ‎

Software plug-in brings new clarity to images of Earth

posted 20 Sep 2012, 07:16 by Mpelembe Admin   [ updated 20 Sep 2012, 07:19 ]

Chinese scientists have developed a software plug-in that generates images of Earth in unprecedented detail. By combining data from two orbiting satellites, the Hong Kong-based researchers say they can produce high-resolution images for monitoring air quality, vegetation growth and a host of other environmental factors that contribute to decisions about urban planning.

The software plug-in was developed by scientists at the Chinese University of Hong Kong (CUHK). It combines data from two satellites to generate images of high resolution and angle variety faster and with greater clarity than ever before.
The team, led by CUHK's Professor Huang Bo developed a method of marrying data from NASA's Landsat satellite, and the MODIS instruments aboard the Terra and Aqua satellites.

Huang says while each of the satellites perfoms a remarkable job, they are not without limitations. Terra takes high quality images of weather systems and their impact on Earth with pixel sizes of 30m by 30m, but can only return to the same location once every 16 days and get data from one angle. Terrra MODIS and Aqua MODIS on the other hand, are viewing the entire Earth's surface every one to two days, and can obtain images from nine different angles although those images are relatively low quality with pixel sizes of up to 1000m by 1000m.

As a geoinformation researcher, Huang Bo, decided to build a unified system that combined the best of both worlds.

The team experimented with two sets of publicly available U.S. satellite data from 2001 - one from the MODIS instruments and the other from Landsat.

From a pair of pictures - one low-quality MODIS image and one high-quality Landsat image - of the same location, the plug-in can establish links between the two. When the researchers feed another low-quality MODIS image of a different location to the software, the plug-in can apply algorithms and models to generate a synthetic, high-resolution image of the new place in seconds.

"This satellite sensor can only provide the data with low spatial resolution. You see this image is not very clear but this one is quite clear. But this one can provide data with frequent coverage, but this one just less frequent coverage. We hope to combine both of them to generate the satellite imagery with both high spatial resolution and frequent temporal coverage," Huang said.

Huang added that while there has been similar research, his group is the first to generate an image that checks the box in all four spatial, temporal, spectral and angular resolutions.

The group is in the process of further developing the prototype, but Huang says his improved satellite imaging technology already has a wide range of applications ranging from military use to environmental detection, and is especially suitable to capture changes in a dynamic, high-density urban environment like Hong Kong.

"It can support the monitoring of the environment, for example air pollution, water quality, monitoring the vegetation growth and status, can monitor the illegal structures, hillfires," Huang said.

One of the first places to test-drive this new technology has been Chinese University's own Satellite Remote Sensing Receiving Station.

Professor Lin Hui, who directs the university's Institute of Space and Earth Information Science, says the new plug-in could improve the quality of data in his institute's research which studies satellite images that shed light on ocean surfaces, earthquakes, air quality information and more.

"When we work with remote sensing, we hope that we can get (any data) from anytime. But we can't whenever it's cloudy. So we capture images that are low in spatial resolution. But since we cannot observe the details, we have to figure out other ways. So if this system can be widely applied, it should be a breakthrough," Lin said.

Huang's group will also try to develop a smartphone application that shows high-resolution satellite images that show air pollution index and land surface temperature.