1st Year Data Acquisition Completed



The first airborne remote sensing data from pilot territories in Cēsis, Burtnieki and Sigulda were collected in the summer of 2015 as part of the SENTISIMULAT project led by the Institute for Environmental Solutions (IES). The data were acquired by using Airborne Surveillance and Environmental Monitoring System ARSENAL.

Sentinel-2 data were simulated by using two of ARSENAL’s hyperspectral sensors: CASI- 1500 with spatial resolution 0.5m/px and SASI-600 with spatial resolution of 1.0m/px. Furthermore, ARSENAL’s high resolution RGB camera and LiDAR laser scanner were used for increased precision of surface cover maps and support for data geometric correction.

False colour image of Burtnieki Municipality pilot territory produced from the hyperspectral data. Image: IES

Data acquisition flights

The flight was carefully planned to ensure that the sun was directly in front or behind the aircraft, thereby avoiding unnecessary difficulties for data pre-processing. The total mission flight time was c.a. 4 hours. Initially collected on ARSENAL’s hard drives, the data was the pre-processed to enable visual evaluation of the data and atmosphere impact. Additional data were collected from Sigulda pilot site.

Data acquisition flight lines in the SENTISIMULAT pilot territories – Burtnieki Municipality, Cēsis Municipality and Sigulda Municipality. Image: IES

The outcome of data acquisition

When the acquired data were collected and pre-processed the mosaic images for further Land Cover/ Land Use (LC-LU) analysis were developed. Mosaic images are created by combining a number of separate flight lines. Three different data types were used to create images that reveal the information about the LC-LU parameters in the studied areas.

True colour mosaic image of Cēsis pilot territory produced by hyperspectral data. Image: IES

False colour image of Cēsis pilot territory produced by hyperspectral data. Image: IES

Digital surface model of Cēsis pilot territory produced from LiDAR laser scanner data. Image: IES

The aim is to use the data to develop a semi-automated LC-LU classification algorithm for Latvia. It would improve the accuracy of the CORINE Land Cover (CLC) mapping.