Abstract

The Thermal Infrared Sensor (TIRS) of Landsat 8 suffers from a stray light issue, where out of field-of-view (FOV) radiance reflects into the optical system and is recorded by the sensors. This is confirmed to be resulting from a defect in hardware of the third order lens. The TIRS-on-TIRS algorithm has been proposed to be an operational correction algorithm. This algorithm has an advantage of simple and easy processing steps. However, no comprehensive evaluation of the TIRS-on-TIRS algorithm has been performed to this point.

This thesis addresses a full evaluation of the performance of the algorithm with two datasets; especially associated with two artifacts related to the stray light issue: absolute radiometric error and "banding" effect. The dataset with truth, MODIS, demonstrates a good performance of the TIRS-on-TIRS algorithm in terms of both absolute radiometric error and the "banding" effect on all situations except for a higher absolute error for desert scenes after correction. The dataset without truth shows good consistency in terms of absolute radiometric error and no worse performance on the worst situations. Residual pattern error was found with band 11, but almost none in band 10. This should be taken into consideration for further calibration work.

Publication Date

9-2017

Document Type

Thesis

Student Type

Graduate

Degree Name

Imaging Science (MS)

Department, Program, or Center

Chester F. Carlson Center for Imaging Science (COS)

Advisor

Carl Salvaggio

Advisor/Committee Member

Emmett Ientilucci

Advisor/Committee Member

Aaron Gerace

Campus

RIT – Main Campus

Share

COinS