Rafal Dabrowski, Agata Orych

Chosen problems with acquiring multispectral imagery data using the MiniMCA camera

View full text in PDF format

Abstract

In recent years there had been a visible increase in the availability of light – “amateur” UAV’s, which can be adapted for remote sensing applications, e.g. research concerning water pollution. One of the many methods used to detect pollutants in water is a method based on spectral reflectance coefficients. Spectral reflectance coefficients can be obtained from imagery acquired in different spectra due to their close correlation to the pixel value. The miniMCA6-channel camera had been specially designed for UAV applications. It enables the acquisition of multispectral imagery data in varied ranges of the electromagnetic spectrum, the bandwidths of which are determined by the applied interference filters. One of its biggest drawbacks is the inability to set exposure parameters for each channel separately. Depending on the transmission coefficient of the interference filters used for each channel, the exposure time for each channel should be different. This can be only adjusted using a numerical value linking the exposure time of the master channel with the exposure time of the slave channel depending on the used interference filter. Therefore the value of a single pixel in the individual spectral channels are often burdened with errors. This phenomenon is most noticeable when acquiring imagery of spectral reflectance standards. Pixels representing the white reference standards on an image acquired by a slave channel are usually overexposed, that is not the case in the master channel, which has a negative effect on the possibility of acquiring reliable spectral reflectance coefficients of investigated samples. The article contains a description and the results of an experiment to determine the optimal exposure time for the miniMCA camera (as a function of light intensity and distance to the sample), to ensure that the digital number of the pixels representing the white reference standards is always a chosen stable value.

Keywords: Multispectral imagery data; data processing; miniMCA; UAV.