This paper presents a method for characterising spatial responsivity of hyperspectral cameras. Knowing the responsivity of the camera as a function of pixel coordinates allows applying a flat-field correction on image data. The method is based on scanning the field of view of the camera with a broadband radiance source, based on an integrating sphere, and combining the captured frames to synthesise a uniform radiance source filling the whole field of view of the camera at the focus distance. The method was compared with a traditional approach where the aperture of an integrating sphere is imaged from a close distance, filling the entire field of view of the camera. The measurement setup was tested with a hyperspectral camera, based on a tunable Fabry–Pérot interferometer. Without the flat-field correction, the average standard deviation of the pixel responsivities across all the spectral channels of the camera was 3.78%. After the correction, the average standard deviation was reduced to 0.40% and 3.25% for the aperture-scanning method and the close-distance method, respectively. The expanded uncertainty ( k = 2) for the flat-field correction obtained using the scanning method was 0.68%–0.78%, depending on the spectral channel of the camera.
No takes yet. Share an insight, caveat, or question.
Kokka et al. (2019) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: